THE SIGNIFICANCE OF PAPYRUS BODMER II AMD PAPYRUS BODMER XIV-XV FOR METHODOLOGY IK NEW TESTAMENT TEXTUAL CRITICISM

Gordon Donald Fee

A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Religion)

August 1966

UNIVERSITY O r SOUTHERN CALIFORNIA

T M I OHAOUATC SCHOOL

UNivKRsmrpANK LMANOELf*. CALIFORNIA 90007

This dissertation, written by

GORDON DONALD FEE

under the direction of h&.-Disscrtation Com mittee, and approved by all its members, has been presented to and accepted by the Graduate School, in partial fulfillment of requirements for the degree of DOCTOR OF PHILOSOPHY

Dean

D*u

AiigagtJjIOJL

Jte^CX

TABLE OF CONTENTS CHAPTER , I.

!
i

PAGE . . . 1 3 12 13 16

INTRODUCTION . . ' She Problem . . . . . .

Statement of Purpose . . . . . . | Importance of the Study

. . . . . . . . . . .

Organization of the Remainder of the Study II. THE PROBLEM OF METHOD IN ANALYZING RELATIONSHIPS BETWEEN ANCIENT MANUSCRIPTS . . . .

18 21 25 29

Codex Slnaltlcus In the Gospel of John . . . . An Historical Sketch . . . . . . . . Attempts at Refinement of Method Codez Slnaltlcus in John 4 . . . Codex Slnaltlcus in John 1-9 . John 1-3 John 5-8 . The End of the Western Text in . . . . . Conclusions . . . . . .

4l 51 55 59 68 71 73 73 83 85 9* 105

. . . .

III, AN ANALYSIS OF THE TEXTUAL RELATIONSHIPS OF P66 Previous Analyses of ?66 Textual Relationships of P66 John 1-5 John 6-7 John 8-9 . . . . . . 11 . . .

John 10-14 . . . . . . . John 15-21 Coneluelone

... . . .....

110 ! 112

. . . . . . . . .

116

TEXTUAL AND SCRIBAL CHARACTERISTICS OF P66 . . . 118 Textual Tendencies of P66 Where It Disagrees with Its Basic Textual Tradition Word order Conjunctions Variations in verb fores Personal pronouns The article before proper nouns Miscellaneous variations The Corrections of P66 The nature of the corrections . . . . . . . ...... 120 122 132 135 144 145

162 167 167

Textual relationships of the corrections 173 Textual characteristics of the corrections 184 THE TEXTUAL RELATIONSHIPS OP P75 IN THE GOSPEL OF ICKE P75 and in Luke 10 P75 and in the Reeainder of Luke Textual Characteristics of P75 and Conclusions . . . . . . . . . . . 192 194 203 213 220

THE SIGNIFICANCE OF P66 AND P75 FOR THE PROBLEM OF METHOD IN NEW TESTAMENT TEXTUAL CRITICISM . . 223

ill

P66 and P75 &nd the Role of External Evidence in Textual Criticise -. . . . . . . ' .... 226 229 233

Singular readings and original text Author's style and original text

The evaluation of variants where internal questions are indecisive P66 and P75 and Early Textual Recension The Neutral texttype as recension Recension and texttype P75 and the Neutral texttype as recension . . 239 ... 245 247 252 . 253 257

Origen and the Keutral texttype as recension P66 and the Neutral texttype as recension

. 259

The Keutral texttype and the "original text" 26l Conclusions: "Reasoned" Eclecticism as a Valid Text Critical Method APPENDIX I. Variation-units and Singular Readings in John * ' 273 264

APPENDIX II. Corrections of P66 Where the Original Text Has Singular or Sub-singular Readings . . 282 APPENDIX I H . Disagreements Between P75 and Codex 286 298

In Luke BIBLIOGRAPHY

Ir

LIST OF TABLES TABLE I Ruber of Disagreements Between Early KSS In PAOB

John 4
II. Agreements in John 4 where at Least Two KSS (not Including TR) Agree Against the Rest . : XH Percentage of Agreements in John 1-8, Based on 320 Variation-units, where at Least Two KSS (not Including TR) Agree Against the Rest . . 17. Percentage of Agreements in John 9, Based on 51 Variation-units, where at Least Two KSS (not Including TR) Agree Against the Rest ....

43

44

52

53

V. Chapter by Chapter Percentages of Agreement Between P66 and Other Early KSS VI. Soee Composite Percentages of Agreement Between P66 and Other Early KSS VII. KS Evidence of the Use of the Article Before in Various Jobannine Constructions VIII. Number of Disagreements Between Early KSS in Luke 10 IX. Percentage of Agreements in Luke 10 at 70 Units of Variation X. Chapter by Chapter Percentages of Agreements in Luke Between P75 and Other Early KSS .... 204 197 196 . 154 88 86

XI.

Singular and Sub-eingular Readings of P75 in Luke Listed According to Types of Variation from Codex 206

TTT. Singular and Sub-singular Readings of in Luke Listed According to Types of Variations from P75 XIII. Disagreements Between P75 and in Luke Where Each Has Important MS Support Listed According to Types of Variation 208 207

vi

LIST OP FIGURES FIGURE 1. A Sample Page of the Collation Used in this Dissertation 40 PAGE

vii

1IST OP ABBREVIATIONS BA BibSac BibTrans BZ CEQ ExpT HTR JBL JTS MS MSS NEB NT NTA NTS NovT RB RevTheoPhil RSV ThIZ TR WH ZNV The Biblical Archaeologist Blbllotheca Sacra The Bible Translator Biblische Zeitschrlft Catholic Biblical Quarterly Expository Times Harvard Theological Review Journal of Biblical Literature Journal of Theological Studies Manuscript Manuscripts New English Bible New Testament New Testament Abstracts New Testament Studies Novum Testamentum Revue Biblique Revue de Thoologie et de Philosophie Revised Standard Version Theologische Literaturzeitung Textus Receptus Westcott-Hort Zeitschrlft fur die Neutestamentliche Wissenschaft

viii

LIST OF TEXTUAL SIOIA The following list includes only the more important uanuscripts cited in this study. The data for other manu scripts cited may be found in Kurt Aland, Kurzgefasste Liste der griechlschen Handschrlftcn des Neuen Testaments (Berlin, 1963).
: ! \ \

I. GREEK MANUSCRIPT Century British MuBeum Papyrus 782 Chester Beatty Papyrus I Chester Beatty Papyrus II III III III II.Ill III II.Ill

A Papyri P5 . P45 P46

266

\ Papyrus Bodmer II

P72
P75

\ Papyrus Bodmer VII-IX

Papyrus Bodmer XIV-XV
\

B. Uncials A C D L V

Codex Sinaiticus Codex Alexandrinus Codex Vaticanus Codex Ephraemi rescriptus Codex Bezae Cantabrigiensis Codex Regius Codex Borgianue Codex Vashingtonianus I Codex Koridethianus ix

IV V IV V V.YI VIII V IV .V DC

 1 13 28 33 38 118 209 565 579 66 700 892 1241

Codex Zacynthlos Codex Laurensls Basel, .. IV 2 Parle, Or, 50 Paris, Gr. 379 Parle, Cr. 14 Paris, CoiBl. Or. 200 Oxford, Auct. D. infr. 2. 17 Venice, 394 Leningrad, Gr. 53 Paris, Or 97 Berlin, Or. Qu. 66 London, Kgerton 2610 London, Add. 33277 Sinai, 260 Family l (Codices 1 118 131 209 1582 2193)

VIII \

VHI.DC
1

Minuscule XII

nil
*

n
XIV IX XHI xucti XI X XII

Family 13 (Codices 13 69 124 230 346 543 788 826 828 983 1689) II. VERSIONS

A. Old Latin a b e Codex Vercellensis Codex Veronensls Codex Colbertinus IV V XII

 t 1 J 1 q r1 eth *m to bc it sa y sy 6 sy 8 gf nB g yP* 1 eyP vg

Codex Palatlnus Codex Brixianus Codex Corbelensls Codex Yindobonensls Codex Saretlanus Codex Rebdigeranus Codex Konacensis Codex Usseriant The Ethloplc version The Armenian version The Bohairie (Horthern) Coptic version The Bohairie Coptic KS edited by Thompson, dating circa 350-375 (see Bibliography) Itala - The Old latin KSS collectively The Sahidic (Southern) Coptic version The Syriac versions The Cureton KS of the Old Syriac version The Mount Sinai KS of the Old Syriac version The margin of the Bare lean Syriac version The Palestinian Syriac version The Peshitta Syriac version The latin Vulgate

T.VI 1 V TO VI VH

Other versional eigla

xi

j
* *

I
CHAPTER I INTRODUCTION In recent years it has often been observed that the proper task of the New Testament textual critic is twofold. The first task is the attempt to discover the original text of the New Testament writings themselves. This is the search for "the New Testament in the ' Original * Greek* (even if the critics' results must be given the uncertainty of quotation marks). The second task Is to interpret the

variations of the text within the context of the history of the Church. This includes, among other things, the discovery and evaluation of the theological Tendenz of textual variation. Although the quest for th2 original text has usually been acknowledged as the first task of textual criticism, the present possibility of the success of such a quest has had varying degrees of acceptance. Some, to be sure, have Intimated that for all practical purposes the ala of the quest has been achieved. Joachim Jeremias, for example,

suggested, "One can say, without exaggeration, that this 1

2 chapter In research is essentially concluded and that we today have attained the best possible Greek text of the New 2 Testament. Manfred Karnetzki, on the other hand, has suggested the abandonment of this quest altogether: Rather than to view the variations as corruptions of a normative, sacred text which needs to be restored, the critic ought to study them as a reflection of living tradition.3 Indeed,

it was in observation of these two trends that Ounther Zuntz had previously bemoaned: After centuries of fruitful work in the field of textual criticism we seem to be faced with an impasse. Many students comfortably pin their faith on the achievements of previous generations; others

The Lord's Prayer (Philadelphia, 1964), p. 7. Prom the translation by John Reumann of Das Vater-Unser lm Lichte der Neueren Forschung (Stuttgart, 1962). Textgeschichte als iiberlieferungsgeschichte," ZNW, XLVII (1956), 170-180. Ibid., p. 170. Cf. Kenneth V. Clark's 1965 presi dential address before the Society of Biblical Literature ("The Theological Relevance of Textual Variation in Current Criticism of the Greek New Testament," JBL, LXXXV [1966], 1-16) where he reflects something of this point of view: "Such scribal freedom suggests that the gospel text was little more stable than an oral tradition, and that we may be pursuing the retreating mirage of the 'original text'" (P. 15). The present study presupposes the validity of the search for the original text and does not attempt a defense against such a position as this. If nothing else, the basic premise of Redaktionsgeschichte should ensure the continuance of the quest. To determine the theological outlook of Luke or Mark, one must first try to determine what they actually wrote.
3 2w

3 !
outstanding scholars among themdecry the search after the original text as chimerical. Between them the two schools leave the critical work to stagnate.** I. THE PROBLEM

If "impasse" and "stagnation" are proper terras to describe the present status of the search for the original text, they are so as a reflection of a deep-seated prob lemthe need for a proper methodology (including a textual theory). In 19^7 Ernest C. Colwell concluded a critique "A new theory and method is

of methodology by noting:

needed. . . . Our dilemma seems to be that we know too much to believe the old; we do not yet know enough to create the new."5 More than anything else, it is this need of a total

theory and method which accounts for any impasse in current textual studies. Since the year 1881 the textual theories for the reconstruction of the text of the New Testament have basically been three:0
i

(l) the adoption of a substantially

!

.
The Text of the Epistles (London, 1953), p. vii.

-'Genealogical Method: Its Achievements and Its Limitations," JBL, LXVI ( 1 9 W , 133. ^Beyond this general statement on the NT as a whole, at least two further theories have been advocated for the Lukan corpus due to the striking character of its Western text. Associated with the name of Priedrich Blass (and later accepted by many others) is a theory of two editions by Luke. See, e.g., The Philology of the Gospels (London, I898), pp. 96 ff. A theory of translation from Aramaic was .

Keutral text,? a theory worked out In Its classical form by

. P. Westcott and P. J. A. Hort by use of the genealogi cal method; (2) the adoption of a substantially Western text, a theory which found Its strongest advocate In Albert C. Clark,^ and which also had rtgenealogyH as a basic method; and (3) the adoption of an "eclectic" text, based on eclec ticism as a methodology, In which ideally no texttype is the basis of our reconstruction, but every variant is considered on its own merits. While each of these theories has had its champions, there can be little question that the currently reigning theory is the "eclectic", a theory which seems to imply that the original text was scattered as pieces of a puzzle in various textual traditions, and that the task of the critic is to find the pieces and fit them together. But as a methodology, eclecticism has taken two basic forms. On propounded by Charles C. Torrey in "The Origin of the Western' Text," Documents of the Primitive Church (New York, 194l), pp. 112-14b; but it was never received with much enthusiasm. 'The terms Western, Neutral, and Byzantine will be used without quotation marks to refer to the three major text groups. It is to be understood that the terms always mean so-called".

'

The New Testament In t h e O r i g i n a l Greek, [Vol. I I ] I n t r o d u c t i o n t Appendix, 2nd e d . (London, 1896).

5
the one hand, there is a thoroughgoing, or "rigorous," eclecticism, such as that advocated by George D. Kllpatrick,10 in which "internal" considerations ideally are the sole criteria. This means preference for a variant which best accords with the author's style, irrespective of - the date and nature of the external evidence which supports the reading. On the other hand, there is a less thorough

going, or "reasoned," eclecticism which seeks a balance between external and internal considerations. According to Leo Vaganay, this form of eclecticism means that there should be no shutting up of the different branches of the science into watertight compartments; verbal criticism, external and internal criticism, all have their parts to play, and they must give each other mutual support, understood in this way the eclectic method seeks a middle way between the two main systems that at present overn the editing of classical and mediaeval texts i.e., wholly external; wholly internal]. The most

"""Western Text and Original Text in the Gospels and Acts," JTS, X U V (1943), 24-36; "Western Text and Original Text in~tEe Epistles," JTS, XLV (1944), 6-65. For a recent example of his method at work, see "An Eclectic Study of the Text of Acts," in Biblical and Patristic Studies, ed. J. N. Birdsall and R. W. Thomson {Freiburg, 9"377~ 64-77. This method also finds expression in his various studies of lexical and grammatical usage of authors of the NT. See e.g., "Some Notes on Marcan Usage," BlbTrans, VII (1956), 2-9, 51-56, 146; "Some Notes on Johannine Usage," BibTran3, XI (i960), 1-5; "Atticism and the Text of the Greek New Testament," Neutestamentliche AufsStze, ed. J. Binzler, et al. (Regensburg, 1903;, pp. 125-137.

6
efficacious method in the field of New Testament criticism borrows from these two schools the best they have to offer.11 While less thoroughgoing eclecticism has been generally acknowledged as the contemporary method, it has been accepted with varying degrees of enthusiasm. On the

one hand, R. V. 0. Tasker, in the "Introduction" to the Creek text of the New English Bible, notes: "The fluid state of textual criticism today makes the adoption ot the eclectic method not only desirable but all but inevitable." K. W. Clark, on the other hand, while acknow-

ledging that the method is "openly embraced," accepts it reluctantly: "it is not a new method nor a permanent one;

it does not supplant the more thorough procedure of Westcott and Hort but only supplements it temporarily. The

eclectic method cannot by itself create a text to displace Vestcott-Eort and its offspring.n l 3 Clark's hesltance, however, is the result of his convictionshared by manythat "Westcott-Hort and its offspring" must in fact be superseded. This conviction

^ A n Introduction to the Textual Criticism of the Hew Testament, trans. B. V. Killer (London, 1937). 12 The Greek New T e s t a a e n t (Oxford and Cambridge, 1964), p. vii. ^'"The Effect of Recent Textual Criticism upon New Testament Studies," The Background of the New Testaraent and Its E3chatoloKy, ed. W. D. Davies and D. Daube (Cambridge,

1954), P. 37.

7
arises from two factors: 1. Vestcott and Hort arrived at their text on the basis of the genealogical method; but as a total method genealogy suffers from two basic limitations. In the first place, as Colwell put it, the "method can trace the tree down to the last two branches, but it can never unite these last two In the main trunkit can never take the last 14 step." The second limitation springs from the apparent "almost universal presence of mixture in [the] manu15 scripts. ^ Vestcott and Hort frankly acknowledged the first limitation: Where the two ultimate witnesses differ, the genealogical Bsethod ceases to be applicable, and a comparison of the intrinsic general character of the two texts becoaes the only resource.10 They also clearly recognized the second limitation, but tried to escape it with their analysis of "conflate readings.* However, as Colwell pointed out, their extension of the argument from the presence or absence of conflate readings to the presence or absence of mixture in a text is not wholly convincing. estcott and Hort therefore used genealogy for one basic reason: to be rid of the "Syrian" (Byzantine)

^Genealogical Method," p. 113.

15

Ibld., p. 114.

3-6rhe Hew Testament in the Original Greek, II, 42.

8
texttype as a serious contender to represent the original NT text. Once they pushed back to what they concluded to be the "two ultimate witnesses," they chose the Neutral over the Western on intrinsic grounds. Insofar as genealogy is a partial method and insofar as it has not adequately answered the problem of mixture, contemporary criticise has sought a new method. Eclec ticism as a method could perhaps well fill this gap. The crucial problem, however, does not appear to arise from eclecticism or genealogy as a method, but from the basic question of textual theory, i.e., by what theory one accounts for the many variations In the existing MSS. 2. Vhen Westcott and Eort chose the Neutral over the Western texttype, they did so on the basis of the following theory of the transmission of the text: Where then one of the documents is found habitually to contain these morally certain or at least strongly preferred readings, and the other habitually to contain their rejected rivals, we can have no doubt, first, that the text of the first has been transmitted in comparative purity, and that the text of the second has suffered comparatively large corruption; and, next, that the superiority of the first must be as great in the variations in which Internal Evidence of Readings has furnished no decisive criterion as In those which have enabled us to form a comparative appreciation of the two texts, (p. 32) Their Judgment that Codex Vaticanus (B) best represents the text which has been "transmitted in comparative purity," is well-known: It will be evident . . . that must be regarded as having preserved not only a very ancient text, but a

9
very pure line of very ancient text, and that with comparatively small depravation either by scattered ancient corruptions otherwise attested or by individualisms of the scribe himself, (pp. 250-251) Hence their basic theory was that the Neutral texttype represented a relatively pure line of descent from the original, and that all other texttypes show varying degrees of "corruption" from this relatively pure line of descent. It was this appraisal of the "neutral" quality of their Neutral texttype which has been the most disturbing factor about Westcott and Hort's theory. In contrast to their Judgment of B, one of the "assured results" of contemporary textual criticism is that the Neutral texttype represents a scholarly recension created in Alexandria in the late third century. . V. Clark goes so far as to suggest that this is "the 18 most influential factor in recent criticism."

^'Such a recension associated with the name of Hesychius was apparently first advocated as early as I808 by J. L. Hug (Einleitung in die Schrlften des Neuen Testa ments); but it was- the revival of the theory by Wilhelm Bousset (Textkrltische Studien zum Neuen Testament [Leipzig, lti94], pp. 74-110) which led to its general acceptance in recent criticism. The theory of recension, although apart from Hesychius, received great impetus in this century when it was openly endorsed by Sir Frederic Q. Kenyon. See e.g., The Chester Beatty Biblical Papyri, Fasciculus I, General Introduction (London, 1933), PP. 14 f and "Hesychius and the Text of the New Testament, Me'morial Lagrange (Paris, 1940), pp. 245-250. What appeared to be the clinching argument for this position came from Zuntz's skillful reconstruction as to how the process probably came about in Alexandria (The Text, pp. 271 ff.).
l8

"The Effect of Recent Textual Criticism," p. 37.

10
It is this "most influential factor in recent criticism" which seems to be the crux of the contemporary problem; for the eclectic method, and its concomitant

textual theory, presupposes the basic incorrectness of the Hortian textual theory. At the same time, however, the

results of the eclectic method, as they are illustrated in subsequent critical texts, show very little significant deviation from the text of Westcott and Hort.1^ The anomaly of the present situation perhaps finds its best expression in the critical text of J. M. Bover, who feels that the Western text attests a pre-recensional form of the second century text, 20 while his own text is decidedly Neutral. With this one may compare the candid admission of Kenyon: "Even if it is an edited text, it

may be a well-edited text; and in the case of all ancient literature a well-edited text is the best that we can hope for."1 As long as our "best" text is admittedly not a

"true" witness to the original text, one can understand the ^ C f . the discussion by Clark, ibid., pp. 29-36. He concludes: "The result of our examination is again to confirm that Nestle's critical text, described by Erwin Nestle himself as 'based on the investigation of the nine teenth century*, as late as 1952 still rests heavily upon Westcott-Hort; that few changes have been made from Westcott-Hort; and that the trend of most recent revision has been a return toward Westcott-Hort"(p. 35). Novi Testamenti Biblia Graeca et Latina (Madrid, 19^3), PP. xxviii-xxix.
21 20

The Text of the Greek Bible (London, 19^9), P. 210.

11 concern for a new method to supersede the old. It Is the

"Hortlan face" resulting from a "non-Hortian" textual eethod and theory which has called for a new method, other 22 than eclecticism, to supersede the old. As it often happens, however, new discoveries frequently call for a re-evaluation of many of our "assured results." In the past decade two important new materials 23 (Papyrus Bodmer II and Papyrus Bodmer XIV-XV) have been

placed at the disposal of the text critic. It is the conviction of the present study that these MSS do in fact call 22 With this statement of the problem one might compare some words of Kurt Aland in "The Present Position of New Testament Textual Criticism," Studia Eyangelica, ed. K. Aland, et al. (Berlin, 1959), pp. 721-722. "What does this finally amount to? Simply that even the modern editions which claim to break new ground still in general present the text of Westcott-Hort, although this is now 75 years old, although since then a mass of new and in part revolutionary discoveries have been made . . . and although the principles of New Testament textual criticism have been remarkably developed. The era of Westcott-Hort, Tischendorf and their contemporaries is not over: we are still within it, as far as the practical establishment of the New Testament text is concerned. . . . This situation is alarming." ^Victor Martin, Papyrus Bodmer II, Evangile de Jean, chap. 1-lh (Geneva, 195b); Victor Martin and J. W. B. Barns, Papyrus Bodmer II, Supplement, Evanglle de Jean, chap. 1^-21, Nouvelie edition augmentee et corrigee (Geneva, 1962)., Victor Martin and Rodolphe Kasser, Papyrus Bodmer XIV-XV, Evangiles de Luc et Jean, 2 vols. (Geneva, 19ol). General descriptions of the MSS are given in the introductions to the editions. Photographic facsimiles of P75 were included with the edition. For P66 the photographic facsimllies did not appear until the corrected Supplement in 1962.

12 for some reassessments, especially of our textual methods and theories. Papyrus Bodmer II (P66) is a codex containing most of the Gospel of John. One folio (containing 6:11-35) is

missing, and there are many lacunae from 14:27 to the end. Papyrus Bodmer XIV-XV (P75) contains John 1:1-14:30 and Luke 3:19-18:18 and 22:4-24:53 with frequent lacunae. The general importance of these two papyri is at least threefold: (l) They both date approximately 200 A.D.,

which makes them the earliest significant portions of the Hew Testament available. This means that they lie on the early side of the half-way mark between the original text and the great uncial codices of the fourth century, Vaticanus and Sinaiticus. (2) They both preserve a consid-

erable portion of the same part of the Fourth Gospel, which makes possible direct comparative study. (3) P75 has, in

addition, a considerable portion of Luke, which has a separate textual history from John. This means that its text may be analyzed with reference to these two histories. II. STATEMENT OP PURPOSE The present dissertation seeks to evaluate the combined witness of P66 and P75 both as to the "revised" nature of the Neutral text and to the question of the original text of the New Testament. task, it proposes: To accomplish this

(l) to offer a refinement of method

13
for Analyzing relationships between ancient MSS; (2) by means of this method, to offer a new evaluation of the textual and scribal characteristics of ?66 and F75 and to suggest the significance of these MSS for ethodology in the search for the "Original" FT text; and (3) to point out that the methodology here advocated Is in fact the contemporary method, and to give reasons for its validity. III. IMPORTANCE OF THE STUDY

Although the chief aim of this dissertation is to demonstrate, by a comparative study, the significance of P66 and P75 for the problem ofraethodology,its greater portion is devoted to the problems of the analysis of these two MSS. The need for such analyses may be briefly summarized. The Text of P66. Because P66 was the first of the two papyri to be published, and probably also because of the intriguing nature of its text, many more significant studies have appeared analyzing its text than that of P75. The majority of these studies, however, appeared within the first two years of its publication, and in their details they all suffer the fault of having used the edltlo princeps, without recourse to the manuscript itself or the photographic facsimiles. The problem here is twofold: (l) Since the publication of the photographs in 1962* the editto princeps has been shown to contain a considerable

14
number of error. These errors are not of such quantity as to Invalidate general conclusions about the original text of P66, but they seriously inpair former conclusions about the corrections to the text. (2) The second portion

of the manuscript (John 14:28 ff.) as first published in 1958, after nost of the studies had already appeared. Again, this probably did not cause serious damage to general conclusions; but it would appear that a re-evaluation containing the full witness of the manuscript is in order. . ~The only aajor study of P66 which escaped this latter weakness was the unpublished doctoral dissertation by Calvin L. Porter, - bet his study, too, did not have the advantage of * the photographs or the improved edition of the Supplement. Since his study was statistically oriented, all of the statistics are now In need of serious revision. Moreover, for the most part the earlier studies appear to suffer from the lack of a controlled methodology. For this reason results were often contradictory. Most of

T h l s was first suggested, on the basis of three - plates In the editio prlnceps, by Howard H. Teeple and P. Allyn Kalker, Kotes on the Plates in Papyrus Bodaer II," JBL, LXXYIU (1959), 148-152. After the publication of the photographs, three independent notices appeared: M.-E. Boismard, H3, III (1963), 120-133; Kurt Aland. "Neue Neutestaaenzliche Papyri II," NTS, X (1963/64), 62-64; and Gordon D. .fee, "Corrections of rapyrus Bodmer II and the Nestle Greek Testament,n JBL, LXXOV (1965), 66-72. 5*4 Textual Analysis of the Earliest Manuscripts of the Gospel of John" (Unpublished doctoral dissertation, Duke university, 1961).
2

15 the studies concluded that the textual affinities of P66 ere closest to Codex Sinaiticus (K), though at least two considered the relationship to be closer to B. 2 " It Is the contention of this thesis therefore, that a re-evaluation of P66 is necessary, and that this re-evaluation should speak to four problems: (l) the

relationship of its text to other Greek manuscripts, (2) the relationship of its text to the "original" New Testament text, (3) the characteristics of the recensional activity represented by the corrections, and (4) the significance of the conclusions of (l), (2), and (3) when compared with P75. The Text of P75. The text of P75 has had only one major study, that of C. L. Porter.27 His conclusions as to

its textual relationships appear to be of special importance to any subsequent work in the discipline of textual criticism. His evaluation, however, was limited to the text of P75 in John. As yet no major analysis has appeared analyzing the witness of P75 to the text of

^Edgar R. Smothers, "Papyrus Bodmer II: An Early Codex of St. John," Theological Studies, XVIII (1957), 43^-441. Cf. I. de la Potterie, "Een nieuwe papyrus van het vierde evangelie," BIJdragen, XVIII (1957), 117-127. This latter article was not available to me; see the summary in KTA, II (1957), 19*. ^Besides the dissertation cited above (note 25), see the important study, "Papyrus Bodmer XV (P75) and the Text of Codex Vaticanus," JBL, LXXXI (1962), 363-376.

16
28 Luke. m e present study hopes to fill this gap, as well as to point out the further significance of this MS as it crosses two distinct New Testament textual histories (Luke and John). IV. ORGANIZATION OP THE REMAINDER OP THE STUDY

The remainder of this dissertation is divided into five chapters. Chapter II takes up the problem of method in analyzing the textual character of a given manuscript. Because of the importance of and its alleged relation ship both with Codex Bezae (D) in John 1-7 and with P66 throughout, a test of the suggested method is offered on in the early chapters of John. Chapter III applies the method of textual analysis to P66. The purpose of this analysis is not simply to discover its textual "relatives," but also to provide a basis for analyzing P66 in terms of the "original" text of the NT. Chapter IV analyzes the textual and scribal characteristics of P66, both of its original text and corrections, with a view to suggesting the significance of the MS in the search for the "original" NT text. Chapter V offers an analysis of the text of P75 in Luke similar to the analysis of the text of P66 in 28 The dissertation by C. M. Martini, "la questione del carattere recensionale del testo lucano del codice alia luce papiro Bodmer XIV" (Pontificio 1st Biblico, 1965), did not come to ray attention until the present study had been completed.

17
Chapter III. Chapter VI draws together the results of the preceding chapters, with special reference to the question of textual theory and method In the search for the original New Testament text.

CHAPTER II THE PROBLEM OP METHOD IN ANALYZINO RELATIONSHIPS BETWEEN ANCIENT MANUSCRIPTS

i i

When a new MS of the New Testament is discovered, there are at least three tasks which, ideally, those who analyze its text should undertake. The first, of course, is to determine its date and place of origin and to indi cate other general features of the new find.1 The second

is the more complex task of locating the new find in the history of the existing MS tradition. which this chapter is devoted. It is this task to

Finally, an analysis of its

text as to its witness to the "original'1 NT text should be undertaken. -

The present study assumes this work to have been done for the two papyri under consideration. Descriptions of the papyri are available in the editio princeps of each, and are not repeated here. A date for both circa 200 A. D. seems to be valid. As yet the only question raised as to the dating is that perhaps they are earlier. Cf. Herbert Hunger, "Zur Datierung des Papyrus Bodmer II (P66),M Anzeiger der o3terrelchischen Akademie der Wlssenschaften, Philosophisch-hlstorische Klasse, 190, Nr. 4, pp. 12-23, who would date P66 toward the middle of the second century.

18

19
Almost all the studies which followed the discovery of P66 and P75 were devoted to the second task. Sometimes these studies indicated generalized conclusions as to the third task, but usually they were limited to the problem of placing the new KSS within the history of the textual tradition. The need for refinement of method in this task should have become apparent from the various studies of P66. All were agreed that Martin was wrong in collating the edltio princeps against Souter's text. But not all agreed as to how the MS should be analyzed. Kurt Aland He was

opted for a collation against Nestle*s text.3 seconded by A. F. J. Klijn.

Against this kind of colla-

tion, Heinrlch Zimmermann significantly warned that readings should be weighed as well as counted;^ and his own lists were an attempt in this direction. Moreover, Klijn's,

as well as M.-E. Boismard's important study,6 viewed P66 in

2

See, e.g., Kenneth W. Clark, "The Text of the Gospel of John in Third-Century Egypt," NovT, V (1962), 23-24. 3"Papyrus Bodmer II, ein erster Bericht," ThI2, LXXXII (1957), 164-168. ^"Papyrus Bodmer II (John i-xiv) and the Text of Egypt," NTS, III (1956/57), 332. 5"Papyrus Bodmer II und seine Bedeutung fur die ; Textgeachichte xles Johannes-Evangeliums," BZ, II, n. F. - (1958), 219. 6

"Le Papyrus Bodmer II," RB, IXIV (1957), 363-398.

20 terns of the later tradition. P66 was accordingly styled "eixed," or "neutral In a non-pure way." J. Neville 7 8 Blrdsall, and later K. V. Clark, who also had F75 in view, objected that the later KSS should be re-analyzed In view of the earlier. But in none of these studies was a clearly defined method forthcoming; and as a result the conclusions were often contradictory. At least three dif ferent approaches in terms of method indicated that P66 has Its closest textual affinities with K.^ But two other

studies indicated that the closest relationship is to be found with B . 1 0 It Is in the interest, therefore, of a clearly defined method in analyzing textual relationships that this chapter is directed. However, because of the importance of in the analysis of P66, and because there Is soae ques tion as to its textual character in John, the method here proposed is applied first to 8 in John 1-9. The first section of the chapter sets forth the justification of The Bodoer Papyrus of the Gospel of John (London, I960), pp. 5-9.
8n 7

Text of the Gospel of John," pp. 18-19.

Thls was true of Martin's edition, where he col lated against Souter, of Aland ("Eln erster Bericht*), where he collated against Nestle, and of the unpublished dissertation by Porter ( A Textual Analysis"), where he " started with a quite different Bethod explained further in this chapter. *See supra, p. 15* n. 26.

21 thle procedure. I. CODEX SIKAITICUS IN TEE GOSPEL OF JOHN

In h i s important study on the o r i g i n of t e x t t y p e s , Ernest C. Colwell concludes with t e n suggestions for f u r t h e r i n v e s t i g a t i o n and c r i t i c i s e . suggestions r e a d s : The ninth of these "The t e x t u a l h i s t o r y of the New Testa-

ment d i f f e r s froa corpus t o corpus, and even from book t o book; therefore the witnesses have t o be regrouped in each new s e c t i o n . " 1 1 A c o r o l l a r y t o t h i s suggestion i s the fact t h a t c e r t a i n KSS a l s o d i f f e r from book t o bookand even within booksas t o the type of t e x t they r e p r e s e n t . Codex V, which makes a d i s t i n c t change froa a Neutral t o a Byzantine type of t e x t a t Luke 8:12 and i s lfestern i n
ark 1:1-5:30, is an exaaple of this kind of "divided"

12
US. Therefore, in the latest aanuals texttype groupings, which both regroup fro corpus to corpus and recognize the "divided" nature of certain MSS, appear as a matter of course.^ Part of the purpose of this present chapter is

^"The Origin of Texttypes of New Testament tenuscripts," Early Christian Origins, ed. A. ikgren (Chicago, 1961), p. 13o. See Henry A. Sanders, The Washington Manuscript of the Four Gospels (New York, 1912J. ^Bruce M. ifetzger, The Text of the Kew Testament (New York, 1964), pp. 213-216; ana J. Harold Greenlee, Introduction to New Testament Textual Criticism (Grand Rapids, Michigan, 19o4), pp. 117-lio. ~~
12

22
to re-examine Codex In John as to the poselbillty of its being a "divided" US, rather than simply "mixed." The secondary character of as a witness to the Keutral texttype, especially in the Gospel of John, has long been noted. Ecrt himself had observed: The Western readings are especially numerous in St. John's Gospel, and in parts of St. Luke's: they belong to an early and important type, though apparently not quite so early ae the fundamental text of D, and some of them are the only Greek authority for Western readings which, previous to the discovery of K, had been known only from the vers ions, i4 In his Die Schrlften des Neuen Testaments, Hermann on Soden made a detailed study of the peculiarities of fc.1^ His twelfth and final "peculiarity" was an analysis of I D agreement, in which he concluded that there was t little significant agreement between them in Matthew, Mark, and Luke. But for John the picture was different. Here he Included a listby no means complete^^where and D agreed against all other Greek witnesses. This analysis tended to strengthen Hort's observation about in John; but it was still considered to be a Neutral witness with a mixture of Western readings.

***The New Testament in the Original Greek, II, 151.

15 l6

I : 1: 917-935.

For example in chapter 4 alone, he includes but five occurrences and leaves out the following seven: v. 11 * oou our; v. 14 be mvev 1. ; . 17 1. ex; v. 27 add ? post etnev; v. 38 1. otTcarctXa; v. 42 uapruptav 1. .

The Western elements In & were also brought into prominence in the work of H. C. Hoskier.1? But his primary

concern lay in separating from as a chief ally. He did note the frequent affinities of with the versions, but did little with its affinities with D. He therefore failed

to provide anything constructive in view of these relation ships . Over the years the various manuals have consequently tended to qualify the association of and by some such statement as: "The type of text witnessed by Sinaiticus

belongs in general to the Alexandrian group, but it also has a definite strain of the Western type of readings."1" But at all times has been considered to be basically Neutral, with Western readings. In 1957> M.-E. Bolsmard offered a study of P66 in 19 John 7-9* in which he indicated displeasure with the prevailing textual groupings. Among other suggestions, he maintained that in John 1-8 had closer textual affinities with than with B. In fact he called one of his textual

17 Codex and Its Allies (London, 1914), 2 vols. I^Metzger, Text of the New Testament, p. 46. Cf. Caspar R. Gregory, Canon and Text of the New Testament (Edinburgh, 1907;, P. 337; and Greenlee, Introduction, p. 116.
19

"Le Papyrus Bodmer II".

24 groups S D (S for S i n a i t i c u s ) .
This proposal of Boismard's has been virtually ignored by text critics. difficult to find. The reasons for this are not

j I
'r

In the first place, he has found little

or no following as to the main thesis of the paper, namely, that the scribe of P66 copied alternately, in sections of from five verses to a half chapter, from an exemplar of one texttype and then from another. Secondly, there is proba-

bly a great deal of uneasiness about Boismard's methodology, since the second of his five newly-proposed "texttypes," whose principle witness is Tatian (J), has absolutely no Greek witnesses (the secondary witnesses are sy 8 0 a b e georg, pers, aeth). Moreover, the details on which his conclusion about

One should note at this point how close Colwell came to this conclusion, before rejecting it, in samplings of variants in John 7. See "Method in Locating a NewlyDiscovered Manuscript within the Manuscript Tradition of the Greek New Testament," Studia Evangellca, ed. K. Aland, et al. (Berlin, 1958), pp. 76b f. His final conclusion that in terms of gross statistics . . . S [] is closer to than to D" in John 7 is worthy of note, inasmuch as this is both contrary to the conclusions of Boismard's coincident analysis, and was based on an insufficient methodological principle in an article whose main force was methodological. It should be further noted, however, that Colwell was using this as an illustration to warn against partial comparisons. This present dissertation, and Colwell himself, in collaboration with Ernest W. Tune, in a later paper on method ("The Quantitative Relationships Between MS Texttypes," Biblical and Patristic Studies, ed. J. N. Birdsall and R. W. Thomson lFreiburg, 1963J, pp. 2532), argues that there is also danger in gross statistics," which frequently tend to distort actual textual affinities.

25 D rests are limited chiefly to the analysis of P66 in chapters 7 and 9 (although this indeed is quite convinc ing). He does offer four important examples at the begin

ning of the paper to support his category S D, but they scarcely amount to full-scale Justification, and could Just as easily be fitted into a scheme which sees as Neutral with Western readings. This present chapter proposes thoroughly to invest

tigate Boismard's conclusion about D.

If in fact is,

in John 1-8, a Western MS with Neutral readings rather than vice versa, then this should be clearly spelled out; for such a conclusion may affect in no small measure what one may further say about its relationship to P66 and P75 The problem therefore now is how to conduct such an inves tigation with proper methodological principles. II. AN HISTORICAL SKETCH

The present attempt to propose a method for analyzing textual relationships is not to be considered a new method; it is rather an attempt to refine some meth odological suggestions which have appeared in various sources. Since studies have already appeared in recent

years in which the history of method has been examined, *

21

See, e.g., Edward P. Hills, "The Inter-relation ship of the Caesarean Manuscripts," JBL, LXVIII (19^9)* 141-159. The greater part of this paper deals with the

21

26
the present interest is eiaply to Indicate those factors which lead more directly to the method here proposed. The single most important factor in establishing textual relationships of any kind is to determine the criteria by which such relationships should be measured. Colwell has suggested that ideally the only proper method is to compare a given MS completely with all other MSS. 22 Until some refinement of computer analysis is available, however/ one will have to settle for a partial method somewhere below the ideal. Nevertheless, the validity of any partial method will be indicated by how well it approaches the ideal. Since the time of Lachmann the most common approach to textual relationships has been to count the number of "agreements in error" or "peculiar agreements'* against an external standard. In the case of NT criticism this

external standard has, until recent times, been the TR. The statistical data from this method usually took the form of "the total number of agreements in variation from the TR."

history of method. Hills' divisions as to what constitutes differences of method seem open to question; and his conclusion in favor of sampling from variations from the TR stands directly opposite the position taken in this dissertation. For a more recent survey of the history, see Porter, "Textual Analysis," pp. 98-104, and Metzger, Text of the New Testament, pp. 179-181. "Method in Locating," p. 757.

27
Although there have been variations of application such as Lachaann's "agreement in error" or Vestcott-Hort's "peculiar agreements," collation against the TB has been the t lee-honored method of establishing textual relationships. By it the NT texttypes were first established; and the discovery of the Caesarean text In this century relied solely on this aethod. In recent years, however, the inadequacy of a aethod using variation from an external standard has been called into question with increasing frequency. In 1945 Ifetzger concluded his summary study of the Caesarean text by asking two important questions about aethod. First, "is It licit to reconstruct the ancient 'Caesarean text* from ofttiaes late documents merely by pooling the non-Byzantine variants?" Second, "is it possible to analyze the textual

complexion of a given document merely by utilizing all variants, large and small?"23 His first question raised the problea of the validity of analyses which failed to take total variation Into account instead of some form of partial variation from an external standard.
2

As he cogently observed:

3 n The Caesarean Text of the Gospels," JBL, IXEV (1945), 486 and 468. This article now appears, with some updating, in Chapters in the History of New Testament Textual Criticise (Grand Rapids, 1903J, PP. ^ - 7 2 . The words enclosed in brackets in the succeeding quotations indicate the changes found in the latest edition.

28
Por obviously it la of slight value in determining family relationship to know only that in a certain area a given manuscript agrees with, say, and ten times in differing frost Textus Receptus. If and should differ from the Textus Receptus in ninety[other] instances, the Keutral element in the given manuscript would be slight indeed, (p. 488) This problea was also noted, and a more thorough procedure pleaded for, by Harold S. Kurphy in his study of Euseblus1 Demonstratlo Evangellca. Metxger's second Question, and one which for the most part has been all too often totally neglected, urges discrimination as well as tabulation of variants. Be noted that the possibility of [mere] chance coincidence among usanuscrlpts in agreeing in small variations (involving, inter alia, word order, common synonyss, the presence or absence of the article, the aorist for the imperfect or historical present) has not been sufficiently taken into account, (p. 489) In a similar vein, Zianermann argued against Aland's tabu lation of P66 that "die Iesarten wollen nicht nur gezahlt, 25 sondern auch gewogen werden. In spite of these objections, however, a count based on variations from an external standard continues to be used. Indeed, E. P. Hills, in his survey of methods, con cluded in favor of tabulating variations from the TH. Although the chief objection to Martin's edition of P66 was that he used Souter's text as a basis for collation,

Eusebius' New Testament Text in the Demonstratlo Evangellca," JBL, 3 (1954), 167-168.
2 e

5 Papyrus Bodsaer II, B p. 219.

29
those who objected continued the same methodology; they nerely substituted a modern TR (Nestle) for Souter. IH. ATTEMPTS AT REFINEMENT OP METHOD

The eteps toward analyses of MS relationships on the basis of total variation rather than agreement in variation from an external standard have been forthcoming only in the past decade. An initial probe in this direction was attempted by Porter in his unpublished dissertation. After a survey of various methods used or suggested in the past, he related of his own: The method of analysis used here is based not upon the calculation of agreements between manuscripts, but upon the calculation of disagreements. The computation of disagreements takes Into consideration the individuality of each MS involved.20 Hie method simply consists of collating disagreements in all of the MSS chosen for analysis, and tabulating the percentage of disagreement each has with all the rest. It is to Porter's credit that he chose a method which compared each MS totally with the others; and his method of first counting disagreements has been utilized

104-105. Porter suggested also that his analysis "rests upon wholly different principles from earlier methods or any heretofore proposed." E. P. Hills, however, had pointed out earlier that the basic and most conclusive method of ascertaining the relationship existing between MSS is simply to count the instances in- which they disagree with one another" ("Inter-relationship," p. l4l). This is precisely the method Porter used, and for the very reason that he considered it "basic and most conclusive.

30 In the present study. But his total method seems to suffer in at least three particulars. 1. One needs a surer guide to demonstrate agree

ments between MSS. Porter, to be sure, recognized that statistical data based on disagreements are only prelimin ary. As a preliminary indication they do have value in pointing out possible close relationships. But the true nature of such relationships can only be demonstrated in a more positive fashion. Porter, therefore, proposed a "second stage," in which he was "concerned with the isolation of those MSS which more or less consistently support the same variant readings" (p. 105). The actual working out of this stage generally took the form of "the agreement of two against the rest." Whereas this is certainly a valid part of looking at agree ments, it would seem to be only a part. For at some point one should look at the percentage of agreement over a total area of variation, and such percentage should be reached only when there has been discrimination before tabulation. The objection here may best be explained by illustration. In John 4, has the following number of

disagreements with these MSS: P66 - 72 P75 - 72 - 70

t :S
D W - 75 - 87

31 Prom these numbers, one may say only that the relationship of to these seven MSS is not very close. However, in I this same chapter of John there are the following number of singular (or nearly Bingular) readings as far as Greek MSS are concerned: P66 P75 D A C W This means - 5 - 6 -2 - 25 - 21 - 1 - 3 - 16 that . (12 have Old Latin or Old Syr support) (12 have Old Latin or Old Syr support)

(5 have Old Latin or Old Syr support) the number of disagreements between and

the other MSS, where at least two Greek MSS support each variant, is as follows: P66 P75 A C D V 42 4l 43 51 40 29 46

One certainly cannot make too much of these numbers, but they do significantly alter the perspective of the disagreements. It would seem, therefore, that an adequate

method of demonstrating agreements is necessary in order to show the relationships which may be involved. Porter, it

has been noted, limited his analysis of agreements almost exclusively to the agreement of two against the rest. And this leads directly to our second objection to his total method.

32
2. One's method must be flexible enough to have

perspective for all MS traditions. The objection to Porter here stems from the limitations imposed by his choice of MSS. In the first place, there may be a fallacy in the

basic assumption as to what constitutes the "earliest" MSS. He is correct in choosing those which actually date from the first four centuries. But he apparently has not considered the possibility that the text found in a later MS may itself date within the earlier period. The case in point, of course, is Codex Bezae. Without doubt, some of its text reflects a later tradition (e.g., where, as in John 4:42, it sides with A and the Byzantine tradition against the early Greeks and most of the Old Latin); but for the most part its basic text has long been acknowledged to be much earlier than the date of the actual MS itself. This failure to reckon fully with the Western tradition could easily lead to partial conclusions, even about the relationship of the early witnesses to each other.27 Moreover, an examination of agreements of "two against the rest" where the entire MS tradition is not in view may also lead to other partial views as to relationships. For example, Porter bases part of his argument for

'Porter (p. l8l) does note that some of the early MSS have readings often called "Western," but he fails to deal adequately with the relationship of the early MSS, and their "Western" readings, to the Western tradition.

33
a relatedness of P66 and on a list of 75 points of variation where they agree against the other early HSS. But a check against the first half of this list (39 vari ants through Chapter 7) indicates that only nine of these are peculiar (or nearly so) to P66 and K. In 14 instances

they are joined (almost alone) by D and in two others by . In seven more they agree with almost all MSS against the Heutral tradition in general. This, of course, does not necessarily destroy the relatedness of P66 and , but it does seem to enlarge somewhat the perspective of that relationship An adequate method, therefore, should be able to keep the total perspective of the MS tradition in view. 3 08 method needs to be able to see relation

ships in sections of a book, as well as over the whole book, d i s is the opposite of the error which Porter correctly argued against, usually MS relationships have

been established by examining test sections. While this is helpful for clues or hypotheses, the more comprehensive efforts to check and establish findings are seldom forth coming (p. 19# n. 25). On the other hand, statistics such as Porter's, which are based on an entire book, may overlook Colwell's principle noted above, that MSS must also be examined section by section within a given book.

3*
One aay therefore use with profit Porter's preliminary step of tabulating disagreements, but the total method must ore In another direction. This direction has been pointed out by Colvell and Tune In their contribution to the Casey Festschrift, "The Quantitative Relationships between MS Text-types." This present study is an attempt to refine further the Bethod there developed. They establish as a basic premise that sound Method should take into account the total amount of variation, not simply the variants from an external text used as a "norm." They further argue that one should exercise discrimination In regard to what is counted. What this means in terms of method is that singular readings are not Included in the counting,2** and only those places of variation are counted where at least two of the MSS involved in the computation agree against the rest. This means of course, that one must exercise care in the choice of MSS to be tabulated and in the extent of text which is included. After the units of variation in the given section are isolated, then the number of agreements Involved among all the MSS at each unit of variation is tabulated. For
2

Their presentation of two tables, one showing percentages with singular readings included and the other without, is sufficient demonstration that they are correct in the exclusion of singulars from the tabulation.

35
convenience this count is finally put Into percentages. Basically, this is the procedure followed in the present analysis, with the following differences from Colwell and Tune: 1. In their further explanation of "discrimination before counting," Colwell and Tune note that they have "eliminated readings which occur commonly in KSS as the result of scribal error or habit, even if supported by more than one KS since such agreement was probable as coincidence" (p. 26). While I agree to eliminate such items as

spelling, I have, on the other hand, included all variations in the counting. The present study suggests that "weighing" may be done after counting. Such a process of

weighing will look at the number and kinds of significant agreements which are involved in the count. This writer discovered that there was a significant correlation between the percentage of total agreement between two KSS and the number of significant readings which were peculiar to the two against all the rest. Conversely, both the number and significance of peculiar agreement between two KSS which did not have a high percentage of agreement was negligible. It was felt proper therefore, as a part of the total method, to analyze such items as peculiar agreements, agreements with or against certain textual traditions, as well as the number and kind of singular readings, in order to gain a full perspective of the relationships one Is

36 .
seeking. 2. The most significant difference between the

present study and that of Colwell and Tune is in the choice of KSS to be tabulated. Their choice was based on an attempt to show relationships between texttypes; therefore, they included the following broad cross-section of MSS: P45 P66 P75 D V C R 565 TR. Since the present interest is more specifically that of determining the relationship of (and later P66 and P75) to other MSS, the choice for tabulation has been narrowed to P45 (where applicable) P66 P75 A B C D W TR. It will be observed that this is simply a list of the major MSS up to the sixth century. Moreover, any singular agree ment between the TR and only one of the others was not included in the number of variation-units counted. * Whereas this may not appear to go beyond the above criticism of Porter, it is believed that it does so for the following reasons: (a) A certain knowledge of MSS is already assumed. The close relationship of P75 and has been clearly

The term "variation-unit" is defined by Colwell and Tune as "referring to. a length of the text wherein our MSS present at least two variant forms; it is that passage in which differences occur," ("Variant Readings: Classi fication and Use," JBL, IXXXIII [1964], 254). One other definition from this article is also used in this paper: "The Nonsense Reading" is "that variant reading which does not make sense, and/or cannot be found in the lexicon, and/or is not Greek grammar" (p. 257).

37 I
I

demonstrated by Porter.3 D is the well-known leader of a quite different type of text. And A has been recognized as being at the beginning of the process called the Byzantine texttype. The selection, therefore, includes at least the

earliest witnesses of the major textual groupings, exclud ing Caesarean, whose text has never been defined in John. (b) The addition of one or more later MSS increases the number of variation-units to be counted, but it always does so in favor of a higher percentage of agreement of all the earlier MSS with one another. Moreover, as a simple

mathematical phenomenon, the percentage of increase is much higher at the lower end of the spectrum than at the higher. One may illustrate this from the findings of the following analysis (cf. Table II, p. 44). In chapter 4 of John, 6l variation-units were counted, based on the MSS chosen for this study. The inclusion of the other MSS of Colwell-Tune's analysis increased the number to 81. At 6l points of variation and D agreed only 10 times, or l6.4; the agreement of P75 and was f>2, or 85.2#. These were the lowest and highest

percentages of agreement among the MSS tabulated. The inclusion of C R 565 and TR increased the agreement between and D to 27, or up to 33.3#. By the same token

3See especially the article, "Papyrus Bodmer XV," PP. 363-376.

38
the agreement between P75 and was increased by 20 up to 72, or 88.8J6. Among all the early MSS the slightest increase in agreement was between D and W; and their number of agreements was increased by 14. It may be granted that 335 agreement is still low, and that the basic agreements are still reflected when the later MSS are added. But the point of view taken in this

study is that the clearer picture among the earlier MSS is afforded by the present selection. The increased agreement

when later MSS are added seems to Indicate that there is a certain relationship which the early MSS have simply because they are early. This is probably significant when other relationships are being sought, but not for the relationship of the early MSS among themselves. The method of analyzing MS relationships used in this paper, therefore, is a combination of Porter's pre liminary step of counting disagreements and the basic method of counting agreements employed by Colwell and Tune. Besides the difference in MSS tabulated, the chief contri bution in "refinement" of method in this study is the attempt toward analyzing the nature of the agreements after the statistical analysis. Before demonstrating the method in John 4, a word is in order as to the method of collation employed, since it may prove helpful for other such investigations. The

39
method received its initial impulse from another article by Colwell and Tune, in which they presented in diagram form the units of variation in John 11 ^o,.'1 it occurred to me

that such a form of collation might prove helpful over the entire Gospel, inasmuch as all units of variation are quickly and easily visible, and the collation of any fur ther MS against the basic collation would be a very simple procedure. If the collation is properly executed, the text

critic may create for himself his own critical edition, against which, ideally, any and all MSS may be collated. A sample page of the collation is shown in Figure 1; a brief description follows. The collation was made on continuous pages (folding sheets) of computer print-out paper, so that all of John and Luke were eventually written out by hand. The present

collation was made simultaneously against Nestle-Aland (25th edition), Tischendorf, and the TR, as well as the facsimiles or editions of P45 P66 P75 A D and W. After

the initial collation was complete, it was double-checked land the MSS collated by von Soden were also added. After some trial and error, the following items were found to be helpful. 1. Aland. The uppermost text is always that of Nestle-

The TR is noted at the far right beneath the

"Variant Readings," p. 254.

40

!?

5 9

-i
_ - *

<> r

*
reading it supports. One say therefore, with any future collation, also collate against these editions. 2. The blocks used for units of variation should be large enough to enable one to write in three lines of witnesses below each variant. The top line includes the HSS used for the analysis in this dissertation. (One may

prefer to reserve it for all uncials.) The second line includes other Greek support. The third line includes Yersions and/or Fathers. If further room is needed to list witnesses, the variations within each unit may be numbered and supporting witnesses listed below. 3Singular readings, many sub-singular readings,

and isolated versional evidence were listed on a line above the main text, so that the main text was reserved for major units of variation, or for points of variation where two or more KSS have a unique reading against the rest. 4. One should be generous with his own script (leaving plenty of space between words), so that later units of variation may be added on the basis of new finds. IV. CODEX SINAITICUS IN JOHN 4 The eethod suggested above was applied to the fourth chapter of the Gospel of John,3s with the following

^Chapter 4 was chosen for very practical reasons. It is the first chapter where D is complete, and one of the only chapters where C is complete. Moreover, P75 begins

J|2

results. 1. table I . The tabulation of disagreements Is found In As noted above, t h i s tabulation Is chiefly useful Since

for pointing up some apparently close relationships.

t h i s type of analysis in John's Gospel has already been made available by Porter, t h i s step in the method w i l l not be used again u n t i l the investigation of the relationship between P75 and in Luke (Chapter V). I t la of Interest to note that, as one should expect, there i s a correlation between those KSS having the l e a s t number of disagreements and the same KSS when they are tabulated for agreements. What cannot be shown from

t h i s step are the kinds of relationship one may expect to find among the KSS having a higher percentage of disagree ments. From t h i s table appears t o have l i t t l e r e l a t i o n

ship t o anything; one w i l l note that i t s t o t a l number of disagreements i s greater even than that of D. 2. The tabulation of the number and percentage of

agreements i n John 4 over 6 l units of variation i s found in Table 11.33 i t w i l l be noted that the relationships

t o have considerable lacunae a f t e r t h i s chapter. Whereas chapter division i s rather arbitrary, t h i s chapter does include two independent pericopes, and perhaps a third, i f one wishes to divide the "harvest sayings" from "the Samaritan Woman." 33in order to make t h i s analysis complete, the c o l lation of the 6 l units of variation for this chapter, as well as the singular readings, may be found in Appendix I .

TABLE
NUMBER CP DISACSEElSiTS BETHEai EARLY MSS IN JCBN h

TR TR
P66* P66 c

P66* P66c

P75 16 * 35 26
-

17 * 36 27 19
-

K* 83 72 70 72 70
-

Kc

c
11 * 27 19 26 23 68 17 * 35
-

68 71
6$

V 51
h9

U6

36
-

51

22
h6

U6 36 16 17 * 83 51 22

18 *

hh h9

37 11 * 37 77 50
m

10 * 5i
U6

P75 Kc A C D V

35 36 72 18 * U6 27 71 19 *

26 27 70 1U * 37 19 65 10 *

7* 1 75 75
68

19 72 19 * li *
26

17 *
mm

70 17 * 37 23 75
h6

87 60
SS

77 68 75 87

50 17 * 68 60

68 66
-

68 51

35 68
SS

12 * 76

7* 1 51

66

12 *

76


*\
*

ON

ON

CM

NO

m
CM

44

ON

$
?"\
CO

&> *
eij
CO

5
f\

-=t

ft
CO CM

3 SS5 5 S tf
CM CM CM f t
1-4

r"
CO

3
CM

CM

I S
CM

r
1

CO

J
V\

ft

in

cn

i ^ ^> > 3 3
NO

*\
CM NQ

I 3

ON

NO .CM

CM JO

3 ft
NO

C M
-

o\

S>
CM

X
OS
1 CM

ON

25

1%
CO CM

-CM

& $
CM V\

$
V\ CM

CM

III s
CO CM CN CM

CM

CK

&

Si
ON

Jt

ON
CM

*
3 CK * CM
NO

i-l

CM

.CM

Os
CM

ON

CM CM

CM 1
CM
ON

^A

-CM

1
NO

8 iO -

CO

&

CM CM

1* s ^
1 NO 1
CM
04

CM

ft

-3
ON

it

ON

f\

ON

CO

VN.

C M

">
NO

-=t

CM 1

ft

CO CM

*> * > ^ ^
NO

*S
ON

JA

CM SO
-=*

CM
ON

ft 3
V\

CO

-3

3 ^ S S
-=t

CM CN. CM CO

CO NO

NO

NO

NO

CM
ON

ON
CO

NO

2 t
NO NO

CM

V\

v S
%

3 <

ft "^ -a
NO

I
CM

s-NO
SO

CM

&
rH

.
CN

co

fc NO S

)2 )2 o T

* v\

I ss
as

lr\ <*\

CM

ft

a as

45
suggested in Table I between P75 and B, P66 and C, and A and TR are borne out by this tabulation. The significant

thing for K, however, is the much higher percentage of agreement it has with D than with any other MS. When D is used as a base, its highest percentage of agreement is also

with.
Colwell and Tune suggest that "the quantitative definition of a text-type is a group of MSS that agree more than 70# of the time and is separated by a gap of about 10# from its neighbors."32* With this definition, P75 and in John 4 clearly fall into the category of texttype, as do A and TR. D and do not fit the first percentage, but they do the second, i.e., they have a 57# agreement, and for the next closest is 34# (*) and for D it is 44# (TR). This would mean that even if one may not classify D as a texttype, each has a significantly closer relationship to the other than with any other MS. This relationship seems strong enough to classify as a basically Western text in John 4. A further look at

the nature of this agreement strengthens the statistical analysis. 3. Of the 6l units of variation in John 4, 16 of

them are the result of two of the MSS agreeing almost alone against all other Greek MSS. There are two other places

3 4 " Q u a n t l t a t l v e Relationships," p. 29.

46
(4:42 and 4:51) where the unit of variation is the result of two sets of two agreeing alone against all others. Thirteen of these 20 instances of singular agreement are between and D, and in the majority of these they have Old Latin support: 4:9 it* D a b e J om. Iou6atot 4:11 4:14 4:17 4:24 D 472 a b e ff 2 J 1 sy 0 K* D om. post

6e ntvcuv 1.

D aur b e e ff 2 J l r 1 . . * D* ff 2 om. a w o v 1.

4:24 fct* D a J r* 4:27 4:27 4:38 4:42 4:45 I

i

* D bo

1. add post

D 1093 a b ff 2 J r 1 sy s - c D .

* D b 1 r* * D D

. XaXtav

1. 1. (or omit) om.

4:51 4:51

S D t r1

A casual check of these agreements reveals that the majority are much more than merely coincidental scribal errors. One seems to be dealing here with a genuine ', textual tradition, supported by two major Greek MSS and often by the Old Latin (especially a b j r 1 ) . \ The only other clear pairing off such as this in the remaining seven instances of singular agreement is between

47
<

P75 and B: 4:11 4:42 4:52 P75 sy8 om.

75 TIJY 1. 75 1.

These, too, are clearly "related" readings, not simply the coincidence of scribal error. It is worthy of note at this point that P75 has now eliminated what once were the three most significant singular readings of in John 4. The remaining four of the singular agreements are of the "scribal error" type and probably are not significant in demonstrating textual relationships (4:15 P66* D bittpm 1. bifm; 4:23 P66* K* 254 124c . ; 4:42

V 80 b f r 1 sy

om. o n ; 4:54

W a b

1. ). Besides these agreements there are the following instances where and D are Joined by a scattering of Greeks against all the rest.
1

4:1 4:14 4:17 4:25 4:46 4:51

add ante bwam 1 . 1. 6 1. xat om. post

Except for 4:1 these are less significant than the former list, but they do point up the number of agreements and D have together against the Neutral tradition.

48
One other significant reading should be noted at this point. In 4:42 there are the following variants: axtjuoaycv * P66 P75 B A C V Byz pi Da Er 565 pc

Vhile the readings of b and 2 are not identical, they do t > sees to represent what one might call "an independent witness to a common textual tradition." It could be

argued, of course, that D merely reflects a scribal error in tens of the reading of P66 et al. But since this same phenomenon occurs in the succeeding chapters of John (see infra, pp. 60, 65-66), it is very likely that they are here related to a common textual tradition. If this be true,

such "agreeaent" will be an iaportant factor in one*s consideration as to the homogeneity of this textual tradition.35 In contrast to these readings, one should also investigate the agreement of * with the Neutral tradition against D. Such agreement appears to be negligible in this

chapter of John. I note the following instances:

^Another reading of a similar nature, but less im portant is at 4:33, where the najority of KSS read post eXcyov. Here K* (as the only Greek) sides with d (against D) e sy0 in omitting the conjunction. D, on the other hand, reads 6e with a b q rl. One wonders whether D, by adding the *wronga conjunction, is witnessing .to a "West ern * tradition which originally omitted it. The fact that it is only a conjunction, where most MSS tend to be quite independent, lessens the strength of such a suggestion.

*9
4:5 4:15 4:21 * 66 75 add ante ** 75 ^66 otepxjuat(-opat) 1. epxaouat (-opat) pot

66 75 C * W L p c b J l q s a

1. ( Byz read ) 4:25 4:39 4:42 66 75 C* W 053 5^5 1. 1.

75 C* L bo b e ^ 1 q rl

175 66 C* V 083 aur a b c ff 2 1 r 1 vg oau

4:51

75 66* A C V (TR )

1.

4:53 4:53

75 C 0125 053 itP1

om. ev ante exetvn om. oTt

75 F66 A C L p e aur a b c ff 2 vg

Something further should be said about this list. First, only two of these (4:52, 4:51) are significant readings, in the sense of indicating possible close textual relation ships. Secondly, in each of these instances except 4:51 the reading of D is also supported by the entire Byzantine tradition and the TH. Moreover, in these remaining eight, where its Greek text nay be determined, the more important MSS of the Old latin version also support the so-called "Neutral" reading. Meat this seems to suggest, therefore, is not so much that is Joining the Neutral tradition, but that D is here conforming to the Byzantine. The fact that there are similar readings where and D read with the Neutral tradi tion against most of the later MSS (4:46 om. ;

so !

4:47

om. ; 4:50

om. xat or 6e) would seem also to

point to such a conclusion. Here, then, we have instances of "early" against "late" readings, rather than Western against Neutral; and has the "early" reading, while D has the "late." It is true that in this chapter there are two places in which fails to join D in what has been called a West ern reading (4:3 add ; 4:49 om. ); but what one is

to classify as Western in this section of John now becomes a problem. For example, what does one call the omission of

in 4:6, where neither nor D join a aur b e ff 2 J 1 r 1 , supported by 565 69 124 788 pc? What this all seems to say is that and D are definitely related, but that the textual tradition to which they belong lacks the homogeneity that is found in the tradition of P75 (or that or D, or both, have suffered conformation to another textual tradition). This is fur ther demonstrated when one investigates the singular read, ings of our chief MSS in this chapter. 4. The one other point at which and D show marked !

similarity, though not agreement, is the number of singular readings each has."' The number of these has already been

given (see above, p. 31). What is significant is that of the 25 singular (or nearly singular) readings of K, there ^ These are also listed in Appendix I.

51
are 12 which have Old Latin or Old Syriac support. Similarly, of 21 singulars in D, 12 have Old Latin or Old Syriac support. The one other MS with a large number of singulars is W, which has 16, with 5 having Old Latin support. This does not say too much in terms of direct relatedness, but it would seem to indicate that and D both are members of an uncontrolled textual tradition and are under the same influence as, or have been influenced by, the older versions. Moreover, if one has been accus tomed to speaking of D and OL agreement as Western, one perhaps should be prepared to do the same with and OL agreement, at least in John 4. It would seem clear, therefore, that on the basis both of counting and of weighing variants, one must agree with Boismard that, in this chapter of John at least, is a Western text. The problem now Is, what is the extent of as a Western text In John? | V. CODEX SINAITICUS IN JOHN 1-9

A chapter by chapter application of our method over the remainder of the first nine chapters of John indicates that a similar relationship between and D exists through chapter 8. At chapter 9 the picture alters completely.

The graphic demonstration of this is found in Tables III and IV.

52
i"V

<

&

$
CO

4
CM

CM

8*

ON

*^

<>

CM CM

f-4

ft R
r-i

CO

NO

NO

\ \

\
NO

co
CM

CM

0\
CO

s
NO

*o

p-v

R NO
o\
CO 1

CM

CO NO

3
1
CO

CO 1

c v\

CM

n
CO

CM

CM CM

NO

-3

3
\ CM

!
I
NO

I
1
NO

NO

CM

NO
CM

CM

CM

cfi
\ CM

-St

-=t

NO

CM

tf
i-i
Vv

CO

w
CM

CO

NO CM

CO NO

CM CM

NO

efW

NO

CM CM

CO CM

Os
CM CM

1f\
CO

CO 1

CO

CM

-5

1
1

in

f\

CM

CM

CO 1

CO
-=t

n
co 1

ON

CO

CM 1

NO

i-l 1

i-l

NO

CO

i
e
Ox

3
n v> v\
CM

-=t

-=t
NO

en co
*

4
CO

3
1 CO

CO

5
Sb
NO

NO 1

NO

) S ^ ^ n 1 s ^ ^ u

V\

53

-tf

r~

09

0\

* & ; 8 & & ! 3

*\

V\

CM

<

CO I

r-=t
CM CO

tt
CM

*n

*n

CM

3
Ox

3
co
CO

3 rt 5
e
NO NO

CM

3
Ox

36

CM

o\
xO 1

i I

I I
-^

i-4 <n
CM

3
en

1 C"l

co

CO U\
CO

n *n <n

xO

CM

CM

\o

CO

CM

NO

xO

3 rt &
CM

IB a*
xO
CM

-^

ON

CM NO

ft
CO
CO 1

CM

S
en en en

en

&:
* P en en

3
I
CM

Cs

en n

!
en en

Os

f-l

3
33.3

3
CM CO

V\ en

Os
CM

Os

&

3
1

CM "UN

Os
CM

* 5

V)

pi .

<

5*
Table III shows the percentage of agreement over chapters 1-8 at 320 points of variation. Because D has a large lacuna at 1:16-3:26, no statistics are included from this section. The significance of for this section will be suggested below. It will be noted that the percentage of total agree ment between and D is somewhat lower than in chapter 4. But it is also true that the percentage of agreement with is considerably lower. The reasons for this are that in chapter 5 temporarily lacks its close relationship with D (dropping to 37.8# at 45 units of variation), whereas happens to be closer to in chapter 4 than in any other of the first eight chapters (dropping to 21.4# in chapters 6 and 7 at 150 units of variation). Over the entire section

I has a 28^ higher agreement with D than with B. t Table IV shows the percentage of agreement in chapter 9 at 51 units of variation. There is no doubt that here g is once again in the Neutral tradition. A chapter by chap ter analysis for the remainder of John reveals that stays in the Neutral tradition throughout, in the sense that it is more closely related to the Neutral witnesses than to D.37 be sure, it still has some Western readings, but

they are now the exception rather than the rule. What is

37Cf. the statistics for John 11 in Colwell and Tune, "Quantitative Analysis," p. 31.

55
Important, however. Is that is never as closely related to as is P75, and often it is not as closely related to P75 as is C or L. For the most part its departure from P75 is less in the direction of D than in that of those wit nesses which compose the Byzantine tradition.^ John 1-3. In this section D has a large lacuna, but on each side of that lacuna and D are clearly related. In 1:1-16, at 9 points of variation, there are the follow ing numbers of agreeaent:

w
TR
-

66 F75 7
-

K* 1

D 2

7 7
-

5 5 7

6 6 6 5
-

8 6 8 6
0

7 7 7 7
2

6 4 4
2 2

P66 P75

7 7 5
1

3
1 1
-

5 3
2

7 5 3 6 6 7 5 4

7
1

7
4 2

tf5
A C D V

6 8 7
2

5 8 7 3
4

5 6 7
2 2

5
-

8 6
-

4 5 4 3
-

0 2

5 8
4 4

6
2

3
-

7
2

3
4

In chapter 13, for example, has a 4l.4 relation ship with B, 43.2with D, 48.3 with TR, and 51.7 with A. Almost all of its readings with D are also shared by A and the Byzantine tradition. C, on the other hand, has a 72.5# relationship with and a 43. l with A; and L has a 69# relationship with and a 4l# with A.

56 In 3:26-36/ at 7 points of variation, there are these agreements (c reads at only one place and is not included): TR P66 P75
TO
-

*
2 1

4
-

P66 P75
*

4
2

5
-

4 6 4
-

3
2 2 2
-

7 4
2

3
1 1 1

5 5 3 5
2

4
2 2

5 6
1 1

3
1
-

4 3
1 2 1

4
2

1 1

4 3 3
-

xc
A D V

3
-,

3 5 3
-

7 3 5

4
1

4
1

4
2

3
2

3 5

5his indicates that on both sides of the lacuna, g is the closest cosgsanion to D, and vice versa. The ex tent and nature of singular agreement against all other Greeks over these sixteen units of variation is similar to that of chapter 4. 1:4 1:13 1:15 3:31 D it syC-P D om. ex2 oa. \evmr 6c uv 1. e v o .

SDb

K* D a b J 1 q

Besides these there are four other significant variants where and D are Joined by a few Greek witnesses against all the rest. 1:3 1:6 * D P66 71 1. ev D* W add ante

57
3:31 D P75 565 a b e ff 2 f 1 r* sy sa cortv 3:32 D 22 28 5^5 pc a b e ff2 j 1 r 1 sy 8 0 om, At only one point (3:3^ om. ) in these sec tions does join what appears to be the strictly Neutral tradition against D. And again is Joined by the strength om. enavi

of the Old Latin (b e f l ) , while D reads with A and the entire Byzantine tradition. One should be prepared, therefore, to regard as the leading Greek witness to the Western tradition in the section 1:16-3:26, where D is lacking. The following readings seem to be significant in this regard: 1:18 1:20 1:21 1:25 1:28 a 6 om. om. xat add post

e 1 sa

W a b e ff 2 l r* a e sy0

om. xat , Br\Qavta 1. 0

66 a b e r 1

1:28 1:32

sy

add post 1.

a b e r* sy 8 0

xaTaPatvov 1:32 1:32 1:34 * e om. 1.

W b e q r1

* 77 218 b e ff2 syS-c (a sa) .

1:47 1:47 2:3

124 aur a b f f 2 1 r* K* a

. . xat l. ctCev

58 . .

a b ff 2 J r 1 ( e l syhraS) otvov ort l. otvou

2:6 i . 2:11

* 13 346 a e r1 arm

om.

* (66* f q) add npurrnv post (6* f q ante )

2:12

245 249 440 1010 aur a b e ff 2 1 bo Q ot

om. xat

2:14

a f q

xat ra xat . xat

2:15

* a b e ff 2 J l q r 1 no t

enotnocv . . . xat 1. xat

2:15 3:1 3*5 3:8

* a e 1 q

om.

TC

K* aur b c f ff 2 1 vg 1. * pc e 1. add xat

aur a b e ff 2 r* syS 0 ante

The full extent of this witness Is lessened somewhat when one considers that lacks at least one Important Western reading in this section (3:6 add oTt and ), and that it Joins the Neutral tradition without Western support in three sig nificant readings (l:l8 TOO ; 3:l6 1. ; 3:13 om.

om. ).

But the nature of many of the

59 readings in the list above indicates that is a member of the Western tradition for the most part. John 5-8. The percentage of agreement for and D in this section is as follows: Chapter 5
K* TR 0

Chapter 6 * D

Chapter 7 K* D

Chapter 8 * D

31.1

40.0

35.2 34.1 52.5 36.1 50.8 34.4 29.9 26.0 23.9 25. 51.1 44.3

30.6 33.9 61.3 45.2 58.9 40.3 19.7 13.1 17.8 20.9 58.1 51.6

54.2 39.6 35-4 41.7 4l.7 43.8 35.4 33.3 37.5 33.3 58.3 52.1

P66*
P66C

35.6 2 0 . 0 35.6 2 0 . 0
30.2

P75

16.3 15.5 37.8 33.3 44.4

lac.
*

28.9

c .

A C D V
i

17.8
lac.

29.0 32.3 28.9 34.2 51.1

lac. lac. lac. lac. 58.1

lac. lac. 46.7 30.0 58.3

37.8 35.6

33.3

31.6 29.5

19.4 33.?

41.7 37.5

As indicated above, the relationship between and D is less in chapter 5 than elsewhere in John 1-8. However,
1

a glance at the above percentages shows that the decrease in D agreement is not in favor of agreement with another MS or textual tradition. The percentage of agreement between and the others remains much the same, and it continues to agree more with D than with the others, though not by as much. An examination of the variants in this chapter

60 Indicates that many of the phenomenon noted in chapter 4 are found here as well, but not in such quantity. D and have only four singular agreements: 5:13 5:18 5*19 K* D* . om. ouv

D 053 a b e f 1 KD a b 1

notet 1. noiei 1.

5:32 tf* D aur a e q sy0

Besides these there are two other readings where t and D reflect a common textual tradition. t 5:2. The name of the pool where the impotent man was healed has three basic variations, with some spelling differences within the three: BnSJaSa 33 b 1 ff 2 D a rl L e A C Byz pier TO f q 75 W 0125 aur c vg bo 66 sa

This is a case of "triple variation," where there is a clear Western, Neutral, and Byzantine tradition. 5:9. After the xat which begins the second clause, a b e sy8 add xat and D ff 2 add the participle . Again, one has agreement in witness without Identical readings. On the other hand, there are only two readings where Joins the Neutral tradition against D, and without the support of the early versions:

61 5:12 5:17 P66 P75 C* L sa P75 1241 om.

om.

Clearly, therefore, is not leaving its relation ship with D in favor of the Neutral tradition. The real cause of this lessened relationship appears to lie in three factors: (l) It will be noted that in this chapter alone,

of the first eight chapters of John, D Is closer to another KS (A) than to K. D appears to have been influenced more by the Byzantine tradition in this chapter than elsewhere, and this accounts in part for the decrease in Its relation ship to . (2) is simply less Western here than else

where. D, for example, reads almost alone with the Old latin at the following significant places: 5:3 5:9 5:13 5:20 5:20 5:37 add napoAvrtxarv post a e v 1. 1. cptXet oeixYixjtr 1. 1.

, on the other hand, has only one significant reading of this type (5:25 i a b t om. xat ). (3) A look at

the 45 variation-units in this chapter reveals that the majority are of the "less significant" type (word order, add/omit the article, conjunctions, pronouns, etc.). More over, within these less significant variants there is a greater mixture of agreement among the early witnesses than

62 one finds at those places which appear to be more signifi cant. It has seemed to the purpose of this study to find reasons for the decrease in D agreement in John 5, be cause it is issediately clear that in chapters 6 and 7 one

has again a relationship similar to that found in chapter 4. The statistics alone seem strong enough to demonstrate this;39 an examination of select readings confirms it. In the following list of variants, and D have singular agreement, or are Joined by a very few Greeks against all the rest. It will be noted that many of the readings (e.g., 6:11, 17, 19) clearly indicate textual relatedness, not simply the results of coincidental scribal errors. 6:3 6:3 6:5 ** 124 565 pc a aur ff 2 1 * D P66 63 71 . 1.

* 6* a aur b c f ff^ 1 r 1 vg > 1.

6:7 6:11

* D 1424 1. & D a b e q r* sy*P xat

1 6:14 D a aur b ff 2 1 r 1 1.

^ T b e sudden increase in agreeaient between P66 and I is the result of a change in P66, not K. (See Chapter t H I * - 94 ff.

63
6:17 * D 6e \

6:17

() D 80 a eyPal

()

1 6:19 6:22 6:22 6:27 D 106 1321 2145 (latt) 1. D 28 42 b c ff2 r* vg D a sy0 1. (or )

1. (or omit) . )

D e ff^ } syc (? pc a b

q r1

6:33 6:37 6:46 6:64

D add post * D 280 (a e) b sy 8 0 * D a b e r* on.

l. etoiv

g D 1604 a b e q r 1

1. ctotv 6:66 D P66 pc aur b c f ff2 J 1 rl post 6:71 7:1 7:3 7:6 7:8 * D om. ante om. add

* D P66 pc aur a b c e f ff2 1 r* s y 8 0 * D G U pc it vg om. om.

* D 047 106 1200 e syC*P

D pc aur a b c e f vgbo syc

OUTTO

1.

7:10 7:12 7:17 7:26

D pc a b e r1 sy8*c

om.

R D P66 33 it vg sy . D P66 om.

D 49 108 a u r a b e f q r 1 vg s y 0 ^ -

64 ;
J

7:29 7*31

D P66 131 D 66 1.

1.

6e ex

ex 6e ex ) 1.

(Byz TR 7:31 7:32 7:32

D a c e ff2 vg sy c# P D P66 pc c e D P66 Q

add 6e post

ot ot

1. ot 7:35 7:37 7:37 7:46 D 249 aur a b c e ff 2 1 rl vg sy 8 0 P66* v l d 69 pc it vg * D P66* b e * 66* om. om.

1. ) I

(Byz TR 7:47 7:48

33 pc a c e tir sa om. D 1.

The demonstration of this relatedness is further reflected in the small amount of agreement has with the strictly Neutral tradition against D. 6:10 6:47 6:58 7:20 7:49 P75 B H N a sy I note the following:

om. or post J om. post om. post om.

P66 P75 B S C V L T

P66 P75 C W L bo

P66 P75 K W L X 33 213 1241 P66 P75 W 33

1.

Again, as in chapter 4, all of the data point to a relatedness within an uncontrolled, or non-homogeneous, textual tradition. Besides the fact that in chapter 6,

65
where K and D have ao many significant singular agreements, they have only a 51# total agreement, there are three other important factors which point to "relatedness within an uncontrolled tradition." 1. There are at least five instances in chapter 6

where and D reflect the same textual tradition, but not with identical readings. 6:1. D 249 2145 b e r 1 J add after , reflecting a textual tradition which placed the feeding of the five thousand on the Western side of the lake, near the city of Tiberias. This same tradition, though not read by S in verse one, is picked up in 6:23 only by K. In place of , it reads , which "Other boats came from Tiberias

results in a reading:

which was near where they ate . . . 6:23. and D alone read a genitive absolute for

the indicative of the rest of the MSS. () (-) 6:25. D ^ b J r sy^

Pier

D and 28 alone of the Greeks change It is

to a form of ( 28 ; D ).

possible for this to be coincidental; but the difficulty

^This is one of the readings selected by Boismard to substantiate his D texttype. See supra, p. 25.

among the versions in rendering here without resorting to the verb "to come," seems to indicate that and D are a part of this tradition. This looks strongly

like versional influence on the Greek at this point. 6:6l. eyvuu . . . xat * () D Rell tCtuv 6e C 6:71. variations:
i

The "surname" of Judas has the following

Ioxapttorou

P66 P75 C W L Q 33 PC f Byz pier TR D aur a b ff 2 J r 1 * sy*111^

There is no textual relatedness between D and at this point, where Codex Bezae apparently is under the Influence of the latin spelling. But * indicate that the name is not a surname but a town from which Judas came, as in "Philip of Bethsaida." What is significant is that here

appears to reflect a Western tradition; for in the four other occurrences of this name in John (12:14; 13:2, 26; 14:22) D alone (joined by e at 13:2) reads . Since it is argued in this paper that is no longer West ern somewhere after 8:38, it seems clear that it is reflecting the Western tradition here, a tradition reflected by D in every other occurrence of the name in the Gospel. 2. The number of instances in which each is the

only Greek witness (or nearly so) to read with the Old

67
latin or Old Syrlac remains at a very high level, has 18 In chapter 6, and 8 In chapter 7. D has 19 In chapter 6, and 6 In chapter 7. readings (e.g., 6:15 1. ; 6:24 Many of these are quite significant aur a c ff 2 1 J vg syc D b ff2 1 r 1

1. evcPnoev [] ). 3. The witness of P66 in these chapters also seems significant at this point. Through chapter 5, there are only four places where P66 might be considered to be pick ing up a Western reading (1:3 . ; 1:28 cv 1. ; 2:11 4:1 CP66*vid3 1. ). add ;

In chapters 6 and 7

there is a sudden increase in this strain (see 6:3* 5, S6; 7:1, 12, 17, 29, 31, 32, 37, 37, 46). What is significant for the Western text here is that along with these there Is also a sudden increase in the number of instances where P66 reads alone, or almost alone, with either s or D. In

chapters 6-7 there are 15 such agreements with K, and 8 with D. * The fact that such agreement should increase at the very point where it3 number of Western readings also Increases, seems strongly to suggest that the textual tra dition to which P66 is here witnessing lacks homogeneity.

P66 6:7, 58, 64, 64; 7:3, 13, 23, 28, 30, 32, 39, 40, 45, 46, 50. P66 D 6:10, 40, 57, 62; 7:12, 14, 35, 42.

41

68
The importance of noting this lack of homogeneity in the Western tradition is that one must broaden his perspec tive for finding this tradition in John. Although D is undoubtedly its leading representative, it is not neces sarily a "pure" representative. If the long addition by D a ff 2 in 6:56 is Western, so also is the of aur 1 vg sy in 6:15. between and P66 (e.g., 6:64 rsapabtovax Moreover, singular agreements

1. o ; cf. a e q) probably

also represent this traditionat least in these two chap ters. The End of the Western Text in K. As noted above,

there is no question that in chapter 9 is no longer a witness to the Western tradition (except perhaps at infrequest readings). This is demonstrated not only by the statistics of agreements (Table IV, p. 53), but also by its sudden lack of singular agreements with D. in chapter 9: 9:19 * I sy add et ante > There are three

9*35 * 3 syS-P add xat ante > 9:40 K* D 63 253 aur b c e f ff 2 1 r 1 ey*5 bo oa. P66, by way of contrast, has six such readings with D, and only one with K, where it is also Joined by W (9:9 om. ). Moreover, there is a sudden decrease in singular readings in (10, with only 2 having Old Latin support); at the

69
same time, there is & like increase of such readings in D (35$ with 15 having Old Latin support). But since this analysis has been using the rather arbitrary device of chapter divisions, the question remains as to whether one can locate the exact place where drops its close relationship with D. In spite of the continued

high percentage of agreement between and D in chapter 8, there are indications that the break takes place within this chapterThe following considerations seem to point

to some place after verse 38 as this point of departure. There are six places of singular, or nearly singular, agreement in the chapter, but they all occur between verses 12-28: 8:16 8:19 8:24 8:25 8:27 8:28 & D sy 8 - 6 (D) pc b ?e . add xat einev post

, pot etnev ouv 1. etncv or eincv

D pc

D 64 aur c e ff 2 1 add rov post D 28 106 add

Beyond this point there are a number of instances where and D agree with the Byzantine tradition against most of the early MSS. The last significant place where this occurs is in verse 38, where It happens three times: D Byz it P66 P75 C L I 1 omit

K* D P5*6 Byz It vg P75 B C V L X X < p 3 3 p c f b o

K D C N Byz it vg P66 P75 V L 13

upuuv omit

Finally, D has 23 singular readings in this chapter, 9 of which have Old Latin and/or Old Syriac support. More over seven of these latter occur after verse 38. on the other hand, has 15 singular readings, with two having Old latin support, but none of the latter and only two of the former occur after verse 38. This evidence, coupled with that of chapter 9, suggests that even though one may not have certainty as to the exact point, the end of as a Western witness in John is circa 8:39 ffThere is one further factor which points to the fact that is a Western text in John 1:1-8:38, and that is the natter of the correctors of K. Without regard to which, or how many, correctors are involved, the direction of the corrections is quite significant. It will be noted from the various tables of this chapter that the correctors in chapters 1-8 always decrease the amount of agreement between and D while at the same time they always Increaseand substantially sothe agree ment between and every other KS. On the other hand, in chapter 9 the corrector increases the agreement with D as well. This clearly indicates that the first hand of in chapters 1-8 is closely related to D and that the direction

71
of correction is almost always away from D rather than toward it. VI. CONCLUSIONS

From the evidence presented in this chapter, the following conclusions are singled out because of their significance for the remainder of this study. 1. It has been argued that a valid method of

analyzing MS relationships must be based on statistics which try to cover the total amount of variation, not simply variation from a given norm. Moreover, a total

method in this regard must also weigh, as well as count, variants. A method on these principles was presented, and when it was applied to the major MSS of the first five centuries, there were the following important results: 2. Although it was not the major emphasis of this

chapter, the statistics alone confirm the very important conclusion of Porter, that the Neutral texttype existed in a relatively pure form in P75 at least by A.D. 200.**2 Not only do P75 and have a consistently high relationship to each other, but it is also consistently higher than any other two MSS have with each other (including A and TR). To speak of P75 as a "mixed" text would seem to press the

^2See especially, "Papyrus Bodmer XV," pp. 374-376.

definition of that term beyond recognizable limitations. 3. Codex Sinaiticus is a leading Greek representa tive of the Western textual tradition in John 1:1-8:38. The significance of its witness at any point of variation in this section of John should have this conclusion in view. Moreover, any further study of textual relationships in John, in which is a part of the consideration, should also proceed with this conclusion in view. 4. Perhaps the most significant thing about the Western character of 8 in John 1-8 is that it points up the lack of homogeneity which exists within this tradition. The facts here presented seem to indicate that there is an uncontrolled tradition to which certain MSS bear common witness, but that this tradition is not fully represented by any single MS or combination of MSS in the way in which

P75 and represent the Neutral. Each of these conclusions is important as we now turn to the examination of P66 and P75 and their signifi cance to the text critic in the search for the "originaln text.

e.g., Metzger, Text of the New Testament, P. 255.

CHAPTER III AN ANALYSIS OP THE TEXTUAL RELATIONSHIPS OP P66 It was observed in the preceding chapter that the discovery of a new MS presents the text critic with a threefold task: (l) to describe the find and to determine its

date and provenance, (2) to locate it in the history of the MS tradition, and (3) to evaluate its role in the search for the "original" text of the NT. The purpose of the present chapter is to attempt an analysis of P66 in terms of task (2). Since such an

analysis must proceed with former studies in view, a brief discussion of these studies is in order. I. PREVIOUS ANALYSES OP P66

Because of the extensive nature of its text and its early date, P66 was hailedand rightly soas the most significant textual discovery for the NT since the Chester Beatty papyri. As a result, it was afforded a considerable amount of attention within the first two years after its

73

74
publication.1 Host of this attention took the form of notices and preliminary evaluations as to its importance. Others were concerned with its witness at various specific readings.2 Our present concern is with the more significant of those studies which attempted to present some form of analysis of the textual data of P66. Before noting these studies, however, a word is in order about the edltio prlnceps itself. Professor Martin should have been, and was, commended for the speed with which he made an edition of the MS available for study. The first edition, which included John 1-14, appeared late in 1956. This was followed in 1958 by a Supplement, which included the fragments of chapters 15-21. With the help of others, further fragments were identified, and a corrected and enlarged edition of the Supplement appeared in 1962.3 Of still greater importance was the appearance of the

In 1957 and 1958 there appeared at least thirtyfive notices and studies. P75, by way of contrast, within the first two years after its publication received the attention of only about seven studies or notices not count ing Porter's dissertation.

See especially, J. Ramsey Michaels, "Some Notable Readings of Papyrus Bodmer II," BlbTrans, VIII (1957), 150154: Robert V. Funk, "Papyrus Bodmer II and John 8:28," HTR, II (1958), 95-100; Edgar R. Smothers, "Two Readings in Papyrus Bodmer II," HTR, LI (1958), 109-122. 3purther corrections to this edition were noted by the co-editor, J. W. B. Barns, in Museon, LXXV (1962), 327-329.

75
entire set of photographic facsimiles with the 1962 edition of the Supplement. The chief criticism of Martin's edition was that he collated it with Souter's Greek Testament. This fault was corrected by a new collation against Nestles text (22nd edition), presented by Kurt Aland.^ The other fault of the

editio princeps was the failure adequately to have noted any of the corrections of the papyrus.5 Because work could not be done directly from the photographs, all studies prior to 1962 contain some errors in detail, none of which, however, seriously impairs their general conclusions. Moreover, all of the studies of the first two yearsand most of the important ones appeared during this tiaewere limited to chapters 1-14 of John. But again, the appearance of the later fragments, although adding further data, did not noticeably affect general conclusions. As noted in the preceding chapter (supra, p. 20), the majority of studies suggested that the closest textual affinities of P66 lay with the Neutral tradition in general, and with Codex Slnaiticus in particular. The relationships suggested by Floyd V. Pllson, based partly on the work of Aland, more or less represent this point of view: "There

4a

Eln erster Bericht," cols. 168-175-

->See supra, p. 14, n. 24.

76 J
is some degree of kinship with Codex Sinaiticua (Aleph). . . there is no striking agreeaent with Codex Vaticanus (). There seems to be some kinship with Codex Bezae (D),

but it is not close." The fact that P66 apparently failed to be an obvious ancestor of any of the later uncials called forth the con clusion that its text was "mixed." Martin, on the basis of his collation, suggested that it occupied a "position intere-diaire . . . par rapport aux principaux manuscrits."7 Aland concurred: "Er fugt sich nicht in die mit dem 4.

Jahrhundert beginnende Scheidung der Textfamilien ein, sondern reprasentiert die fluktuierende 'Variantenmengung1 der Zeit davor."8 In a similar vein, A. P. J. Klijn, who approached the KS with the question of the origin of in view, con cluded that "in P66 we are again Cas in P45] dealing with a mixed text-type."9 In his Judgaent, however, it is less !

"mixed" than P45, and his final conclusion was that "we find on the whole a relatively ssall number of non-neutral readings: thus P66 may be called neutral, in a 'non-pure
i

A Kew Papyrus Manuscript of the Gospel of John," BA, XX (1957), 61. ^Papyrus Bodmer II, p. 1^9.
8n

j
i

Ein erster Bericht," col. 179-

^"Papyrus Bodmer II," p. 331.

77
way, like the witnesses of this text such as cc, C and L (P. 333). It is probably this problem of "mixture" which has made the analyses of P66 appear to be so contradictory. has already been noted that at least two early studies found the closer affinities of P66 to be with rather than (supra, p. 15). M.-E. Boismard10 and W. Hartke 11 took a quite differ ent stance to account for the "mixture. Both posited an hypothesis that the scribe had two Vorlagen from which he made his original copy. Hartke, on the one hand, suggested that the scribe copied from a basic text (an archetype of ) , but with constant referral to another text which had been imported from Rome. Boismard on the other hand, con cluded that the scribe followed one Vorlage (a Neutral text close to B) and then the other (a Western text similar to K ) in clearly defined sections. Although Boismard*s study was limited to John 7-9, he attempted to support his hy pothesis by noting ten readings throughout John 1-14 where he considered P66 to have conflations of readings from the two Vorlagen. He concluded his study with a list of It
N
I

10

"Le papyrus Bodmer II."

**"Bemerkungen zu Papyrus Bodmer II," Forschungen und Fortschritte, XXXII (1959), 122-126. This article was not available to me. Cf. the summary in Porter, "A Textual Analysis," pp. 142-143.

78
forty-nine readings previously unknown in the Greek MS tradition, for which he elicited support from the versions and Fathers, especially Tatian. A study by H. Zimmermann12 attempted to note the significance of P66 for the history of the text of John. A part of this study investigated the relationship between P66 and K. He based this relationship first of all on some

common readings, but his major conclusion was that P66 shows substantially the same characteristics which consti tute the peculiarities of K. He concluded that, like K, P66 is chiefly Neutral, with a profusion of Western read ings, often under the influence of the OL. In ?66 one has

"damit zeichnet sich eine Entwicklung ab, die etwa 200 Jahre spater im Sinaiticus noch weitaus starker ausgepragt 1st" (p. 225). On the basis of his conclusion that has a basically Neutral text after the fashion of K, Zimmermann further concluded that the essentials [Hauptsache] of the Neutral text existed in Egypt before 200 A.D., and that it can no longer be maintained that this texttype is the product of a fourth century recension (p. 225). C. L. Porter's unpublished dissertation presented a significant attempt to define the relationships of all the early MSS to one another. Since he had the advantage of

12

"Papyrus Bodmer II."

79
these earlier studies, and since his analysis was attempted along new lines, several of his conclusions about P66 are worth noting. The first part of his study offered a table showing the percentage of disagreements between P66 and the other early MSS. For the major MSS the disagreements with P66 were as follows: P45 P75 W 55.2# 40.8# 57.3# 40.5# 47.9#

But since this table pointed out no significant close relationships, Porter turned to an analysis of the agreements of P66 with one of the other early MSS against the rest. P66 had more of this kind of agreements with (79 Instances) and W (62 instances); he therefore concluded that P66 Is more closely related to these two than to any other MS. (There were only 7 such Instances of "agreement" On the one hand, he noted: "The

with P75, and 11 with B.)

text of P66 is not a part of the history of the text found In Codex Vaticanus," nor is there any "clearly . . . close textual relationship" between P66 and P75 1 ^ On the other

hand, "the text of P66 is a part of the same textual stream out of which Codex Sinaiticus flows" (p. 150). His final conclusion, therefore, was that "the text of P66 is not a

^ " A Textual Analysis," pp. 152-153.

80 'mixed1 text. Its text is best described as one which lies in the higher reaches of the stream from which Codex Sinaiticus and the Washington Codex later derived. " ^ Besides these attempts to place P66 in the history of the MS tradition, there have appeared three other studies which tried to analyze some of the individual characteristics of the text with a view to indicating its significance in the search for the original text of John. The first of these studies, and one which broke new ground in the analysis of the MS, was J. Neville Birdsall's Tyndale New Testament Lecture for 1958.15 He first

reviewed some of the former studies and rather sharply criticized them for discussing and defining "such early evidence as this [P66] by standards of later witnesses" (p. 7 ) . Especially singled out for criticism were Klijn's

terms "less mixed" and "neutral in a 'non-pure' way." Birdsall then offered a brief analysis of the text of P66 in John 10-11 in which he concluded that textual analyses of such early MSS as P66 and P45 which are based on relationships with later MSS lead only to an impasse. "We must emphasize . . . that in fact the papyrus texts are not contaminated varieties of the later texts which we know already or mixed texts made up as it were of later texts"

^Ibid., p. 154. Italics not in original. 15The Bodmer Papyrus.

81

(p. 10). He then proposed a new method for assessing P66 by which its "worth" and its "nearness to the original" may be evaluated. This method was based on grammatical and exe-

'

getical criteria derived from various studies of Johannine grammar. On the basis of such criteria Birdsall concluded "that the papyrus probably preserves a number of original readings but also displays a marked tendency to smooth over certain harshnesses in the original text" (p. 13). His final conclusion was that "in an acceptable sense . . . this is a very mixed* text. It is a mixture of good and bad, of primitive and recensional" (p. 17). Along similar lines the present writer offered an analysis of the many corrections in P66. 1 " It was sug-

gested that the significance of the corrections lay not in the kinds of textual tradition with which they tend to agree, but rather in the kind of scribal or recensional activity which they appear to demonstrate. The conclusion of the study was similar to that of Birdsall, namely, that the direction of most of the corrections was toward a smoother or easier text, hot necessarily toward a certain textual tradition. E. C. Colwell offered a further study along this

"The Corrections of Papyrus Bodmer II and Early Textual Transmission," NovT, VII (1965), PP. 247-257.

lo

82 line in which be analyzed the scribal habits in the early papyri by indicating the kinds of "editorializing" one finds in their many singular readings.*7 Colwell's major conclusion about P66 is that the scribe editorializesas well as copies"in a sloppy fashion" (p. 37). All three papyriP45 266 P75"show that scribes made changes in style, in clarity, in fitness of ideas, in smoothness, in vocabulary" (p. 382). But in contrast to P75* where "the

scribe*8 impulse to improve style is for the most part defeated by the obligation to Bake an exact copy" (p. 386), the scribe of 266 is careless and ineffective: "He uses up

his care, his concern, in the production of beautiful let ters" (p. 382). The major concern of the next chapter is to Indicate the role of P66 in the search for the original text of John along the lines laid out by these last three studies. How ever, before that analysis is attempted, this present chap ter offers still another look at 266 in terms of its textual relationships. The chief reason for such an analysis is not simply to look for a "way out" of the ambi guity of the conclusions of previous studies, but to offer a point of departure for the investigation of textual characteristics which follows.

^"Scribal Habits in Early Papyri: A Study in the Corruption of the Text," The Bible In Modern Scholarship, ed. J. Philip Hyatt (Nashville, 1965), PP. 370-3^9.

83 I
I

H.

TEXTUAL RELATIONSHIPS OP P66

In the following discussion of P66, at least three conclusions from the preceding chapter are of importance: 1. The affinities between P75 and are such that one usay now speak of the existence of the Neutral texttype in a relatively pure form at least by the end of the second century. This does not necessarily mean that this texttype is either "neutral" or nnon-recen3ional," since it may have developed in the second century. But Aland's expression about a "4. Jahrhundert beginnende Scheidung der Textfamllien"1^ Is definitely not true of the Neutral texttype. Furthermore, the existence of a clearly-defined texttype at such an early date means that it is not necessarily incor rect to speak of any other known text as "mixed as long as it is seen to be "mixed" either toward or away from this texttype. 2. Codex Sinaiticus is first of all a divided text !

in the Gospel of John. In chapters 1-8:38 its text is basically Western; thereafter its text is basically Neutral. Furthermore, in terms of its basic witness in each section,

^"Ein erster Bericht," col. 179. Aland has contin ued to maintain this position in spite of the evidence of P75. See especially "The Significance of the Papyri for Progress in New Testament Research," The Bible in Modern Scholarship, ed. J. Philip Hyatt (Nashville, 19b5), pp. 336-337. For the discussion of this entire question, see Chapters V and VI of this study.

it ie probably correct to call It a "mixed" text. While this Is easily to be seen In its Neutral section, where it is a secondary witness to the text of P75 and B, the ques tion of mixture is more difficult in its Western section, where Itself is the earliest Greek witness to this tex tual tradition. The question remains, by what standard one

measures "mixture." 3. The tentative conclusion that the Western text represents a non-homogeneous tradition, composed of read ings found in various scattered witnesses, reflects a proposition formerly made by E. C. Colwell In his study of the origin of texttypes: "The so-called Western text . . .

ie the uncontrolled popular text of the second century. It has no unity and should not be referred to as the 'Western text. ,nl 9 He also noted that any "texttype is a process,

not the work of one hand" (p. 136). If this be true of the Neutral, it is a process which had, for all practical pur poses, culminated by the end of the second century. But it does seem to be true of the Western and Byzantine tradi tions. Therefore, it Is probably valid to speak of a "mixed" text in the Neutral tradition. But one Is not to consider the mixture as from another clearly defined texttype. It is rather a mixture of readings, some of which

are a part of the Western tradition, and others a part of

^"Origin of Texttypes," p. 137.

85

the process which later emerges as the Byzantine tradition. A chapter by chapter breakdown of the percentages of agreement between P66 and the other early Greek MSS is found in Table V. Examination of the Table indicates

that ?66 is clearly a member of the Neutral tradition in chapters 1-5. In chapter 6 a shift takes place toward

in particular and the Western tradition in general. This remains as a pronounced feature in chapter 7 Thereafter, P66 reflects no clearly defined relatedness, either to a single MS or a textual tradition, although in general it is closer to the MSS of the Neutral tradition than to either D or the Byzantine tradition (A, TR). The clear indication of these general statements will be found in Table VI, where some composite percentages of agreement are listed. An examination of the nature of the relationships within these sections further substantiates the statistical data. ; John 1-5. In this section P66 has its closest

relationship to the three major Neutral witnesses, P75 C. The reason for its closer percentage of agreement with C 20 It will be noted that the percentages are only tabulated through John 14. There are several reasons for this. In the first place, P66 is fragmentary in the re maining chapters of John. Secondly, P75 is entirely lack ing after chapter 14. Finally, only and of the early uncials do not suffer major lacunae in these chapters. The analysis of P66 after chapter 14, therefore, must be more specifically indicated, such as in the discussion below of chapter 19, pp. 113 ff.

86

*-

-=f

CO

CK

S $ 5 ; 3 $ ; S : 3 3

CM A

CM

CO

{ 3 $ & 8 . 3 3 3
A CK

i
2

a .

S.
CA

co

NO

<

CK

co

CM

CK

^ ^ ^ ^ , ,
f CO NO CO CK CM -=t CM

CM

y*
CM

?
..

i 4
CK

c e o

$ i t t i d $ $ * ^ 3
4 U\ VO CK CO

VA * t -

1
r

V\
i H N 0 C

$
r 3

CM O C

ti

e^

s
K

so
N 5 I - 4

.
XA
NO XA NO NO

CM NO

4
CM

.
4
NO

.
CM

CO

CO

C K r i X A C M N O

ft

so
XA

\
CO XA <A

4
CK CO NO NO

3 NS
CM

m
CO A CA CK NO

XA NO

NO

NS

S
CO CM NO CO CM NO

XA XA

NS
A *

CO

s
CO XA CA

CM NO CO CM NO

XA CM NO CO CM

CO CA A

i-i

co
-3 CA

<A

NO CK

CO

^ ^
CM <A NO CM CO NO XA CM A NO

XA

tt

ft
XA

NO CA

'

NO

t-

4
6

A NO

*>

fc

A.

as

as

87

ON

CO

CM

NO

NO

\A

CK *A Os

CM

C A
NO

3

W
CM

^
-^ NO

s 3
\A

C--=t

3 3
CN CA

UN

NO

*A

3 ^
NO
ON CO CA

CA

i-i
NO

NA
ON

,3 3 3
C % M .
C M

3 *
0 \

* 1

.
CO

>
CM

CM

^ 2 3 3

-=*
O

3
O O

?A

3
0

2
CM CM

CO

CO A

CO

CA

NO

CO

CO

3 3 3 3 3
CA CO ON

XA XA

CA

CO CA

-=t

s
VO NO fW

NO Ox XA

CM

-=t
CO XA NO CM CM V\ XA XA CM CM CM

3 3
n o.

^
NO XA

3
es

CO CM

CO

>0
XA

^
CM

NO
0\ XA

NO

3 3

XA
I-

3
CO CO
ON

CM

XA CA CA CA

CM CM

ON

NO

S>
(L,

CM Ox CA A . 3
ON

Ox -=*

CA A A <A

$ 3 3 3 3 , 1
as
<

3
Q > S

88
TABUS VI SOME COMPOSITE PERCENTAGES OF AGREEMENT BETWEEN P66 AID OTHER EARLY IBS NOTEi The f i g u r e s represent nuriber of varia$ion-urn\ts/nuriber of agreements(percentage of agreeaent) John 1-5 P66* P66c 179/91* (52.5) 178/123(68.7) 179/116(65.2) 179/58 (32.U) 179/82 179/89 (16.8) (h9.9) John 6-7 P66* 123/550*1*.7) 112/39 (3l*.8) 123/51(1*1.5) 123/70(56.9) 123/69(56.1) 3i*/l9(55.9) 38/ll*(36.9) 123/50(1*0.7) 123/1*9(39.9) P66C 123/65(52.9) 112/52 (U6.h) 123/59(1*8.0) 123/67(5U.5) 123/78(63.1*) 3l*/21(6l.8) 38/16 (U2.1) 123A6(37.1*) 123/56(1*5.5)

TR

179/92 (51.1*) 178/116(65.2) 1799(6.9) 179/65 (36.3) 179/85 0*7.5) 179/83 0*6.1*) 112/77 (68.8) 122/35 (28.7) 179/91* (52.5)

P75
* c A C D

112/79 (70.5) 122/35 (28.7) 179/92 (51.!*)

John 8-Ui
I

John 1-12* P66C P66* 663/315(1*7.5) 51*7/280(51.2) 663/33l*(50.U) 662/2?5(l*l*.6) 662/311(U7.1) 537/21*5(1*5.6) 309/l50(U8.5) 60l*/235(38.9) 662/298(U5.0) P66c 663/3l*l*(5l.9) 5U7/30l*(55.6) 663/35K52.9) 662/295(l*l*.6) 662/328(149.5) 537/26U(lt9.2) 309/162 (52.U) 60i*/232(38.1*) 662/321(1*8.5)

P66* 361/168(U6.5) 257/125(1*8.6) 36]/L7l*(l*8.2) 360/160(1*1*.!*) 360/157(1*3.3) 32l*/ll*3(l*U.l) 159/59 (37.1) 359/1500*1.8) 360/155(1*3.1) 71/26 (36.6)

j TR
.*#

361/185(51.2) 257/129(50.2) 361/175(1*8.3) 360/l7C(U7.2) 360/168(16.7) 32l*/i5U(U7.5) 159/67 0*2.1) 359/15KU2.1) 360/173(1*8.1) 71/30 (1*2.3)

:
; **

A
c

w
Pl*5

89
is easily explained. P66 fail to read with P75 at six significant variants where they agree almost alone (1:13; 4:11, 42, 52; 5:11, 1 9 ) . 2 1 The closer relationship of P66 to P75 than.to is also easily explained. P75 and have their least close

relationship in John 1 (68.4$), and where they disagree, P66 more often sides with P75 than with B. There is scarcely any significant relationship between ?66 and the Western MSS in this section. There are six instances where P66 and read alone (or almost so) among the Greek witnesses: 1:28 P66 K* a b e r* ev

1. ev 1:33 2:11 4:23 5:6 5:25 P66 add rw ante

P66* (K*) f q add . P66* R* 254 . om.

P66* 053 1321 sa e P66 K*

om. ot ante

But of these only 2:11 is significant in terms of estab lishing textual relationships. The final three are

21 Whether P75 or P66 C et al. are the best repres entatives of the Neutral tradition at these points is a further question which is not at issue here. (However, see infra, p. 121, n. J.) Furthermore, it is my observation that Codex C deserves a closer examination in John as one of the "purer" representatives of the Neutral texttype. This is especially true in the chapters where P75 Is lack ing. C here appears to have a much closer relationship to than does K. See supra, p. 55, n. 38.

90
corrected either in P66 or k, or in both, and each instance probably reflects independent agreement of a "scribal error" type of reading. There are two instances of singular agreement with D, 4:15 5:18 P66 D 6** 1. Cut ot

P66 D syc

1. and a single instance of agreement with W and the OL (1:17 add 6c). In addition, P66 has two readings where it is

Joined by a combination of Western witnesses:22 1:3 4:1 P66 K D l p c . ev

P66*vld D 565 pc itP1 vg sy 1.

Only two (2:11 and 4:1) of this entire group of readings are of any significance in terms of textual relationships. The lack of a higher percentage of agreement with the Neutral MSS, therefore, does not appear to be due to an appreciable mixture of Western readings. It seems rather

to be the result of two other factors: First, P66 is not

^The reading of for P66* in 4:1 is not self-evident; it is not so noted in the editlo prlnceps, nor in the articles calling attention to corrections missed in the edition (see supra, p. 14, n. 24). But the correc tion seems quite certain. One may observe how unlike every other kappa on this page is the kappa of the ". (Note also the kappa in the xc" on the following page in v. 11.) Moreover, the downstroke of what is now a kappa is identical to the iota of " directly beneath it.

91
closely related to any Individual MS of the Keutral tradi tion. It was noted above that it fails to follow P75 and

in readings where they show a very clear textual relation ship. Secondly, and more significantly, tGS varies fro the Neutral tradition in a considerable raaaber of readings of the type which are often supported by the bulk of the Byzantine tradition and which frequently find their way into the TR. The best indication of this is seen by exam ining the instances in John 1-5 where P66, supported by several or the majority of later witnesses, differs from P75 B: (In each Instance the reading of P75 and support

ing witnesses is given first and that of F66 and supporting witnesses is found below.) 1:13 1:32 1:35 1:46 1:49 2:15 P75 * 69* 2*7 P66 D C V L pi TS > 75 A C L 083 pa oei P66 Byz TR 75 L 28 66 A C 083 Byz TR 75 P66c L 33 579 66* Byz TR 75 A L 1 33 579 PC ?66 X Byz TR

ei ct

75 P66c L 083 33 579 PC 66* Byz TR

3:20 omit
4:5

P75BKA063ByzTR

o n 66 3 63 L 33 1241 pc 75 083 Byz TR ouP66CDWLN6pa

4:11

omit 6ttnoct 6itnon

P75 sy8 P 6 6 A C D W L e T 0 8 3 B y z rell TO K P75BKADLT083P P66 X Byz TO

4:14 4:30

75 pier 66 A pm TO C D be 75 C L pc 66 A D 083 Byz TO 75 A C L 083 Byz TO P66 D A pc 75 66 A C Byz TO tt*Db 1 r 1

4:39 4:42 4:42

XaXtav ov

4:50 4:51 4:52 4:53 4:53 5:11

P75 A C L 083 pc 66 D Byz TO P75 L 083 pc P66 A C ( O ) Byz TO

omit xax annrretXav aimuv

P75 P66 & A C D W L rell TO P75 8 A D V L TBjz ffi 66 C 0125 pc

P75 C pc cv P66 A D L Byz TO 6e P75 A P66 C L pm omit D Byz it TO omit P75 1241 66 A C D L N 8 Byz TO

5:17 5:19 5:37

P75 66 D A rell TO 75 L pc 66 Byz TO D

It will be readily observed that the majority of

93
these are of the "less signifleant" type of variation. Two Involve the addition/omission of the article with proper names; two involve the addition or alteration of conjunc tions; two others involve the attraction of relative pro nouns; in three the subject has been added; one is a change from the future indicative to the aorist subjunctive in an emphatic negation; another adds ev with the dative of time; another is the addition of the possessive pronoun; and three reflect alternative words or spellings for similar expressions. nils does not mean, of course, that P66 is a "Byzan tine" MS. In chapter 1 alone there are 28 places where P66 reads with P75 against the Byzantine tradition, and many of these are significant variants (1:18 om. eortv; 1:27 1. ; 1:37 ; 1:41 . ; 1:27

oa. ; 1:30 1.

position of ; 1:39

1. ; 1:49

. om. a t ). r

1. xat ; 1:51

But the foregoing list does indicate that the greatest amount of mixture in P66 is of the "Byzantine type" of reading. It will be further observed that the great majority of readings in this list are clearly secondary to P75 (on the basis of the criterion, "Which reading best explains the existence of the other?"). And almost always the varia tion is in the direction of a smoother, easier, or fuller

text. A clear example of this kind of variation, where P66 has a "Byzantine type" of reading but Is not necessarily related to the Byzantine MSS, is 5:11. The 6e of P66

C V et al., appears to be a half-way house between the probably original of P75 A and the full Byzantine resolutionthe elimination of the relative or substantival article altogether.23 Apart from chapters 6 and 7 (and to a smaller degree chapters 11 and 12), this kind of "mixture" appears to be the chief reason in P66 for variation from the Neutral tradition in the remainder of the Gospel of John. John 6-7. As suggested in the preceding chapter

(supra, p. 62), there is in P66 a definite increase of readings from the Western tradition in these two chapters. The clear indication of this is the sudden increase of Instances where P66 reads alone (or nearly so) with either

3The choice of as original rests chiefly on the canon ardua lectio potior. "John" never elsewhere uses the relative in this manner, and he rarely so uses the article. The only instance of with , if P75 W may. be trusted, is in 5:17. (Here.P66 re11 com pensate by adding the subject , a form which occurs but eight times in John". In every other occurrence of in the gospel, the verb stands first and is followed by the subject, when expressed. Therefore, a scribe would scarcely change to (even by error), and the addition of either would be wholly out of keeping with Johannine style. See the more detailed discussion of this variant in Chapter VI, pp. 240-241.

95

or D, or with both together, against the rest of the textual tradition: 6:7 6:58 6:64 6:64 ?66 add . P66* * tva>v 1.

P66 1. 66 a e q napafctCovat

1. 7:3 66 28 185 sy ot 1. ot 7:13 7:23 7:28 66 544 q 1. . |

66 add ante 66 544 1.

7:30 66* saP* ot 1. 7:32 7:39 7:45 7:46 66 sa bo vept 1. . 1.

66 157 249 aur c ff2 1 q 66 e

1

. (D om.

66* * sys

; Byz TR om. ; P66c P75 W L pc omit clause) 7:50 P66 f (e q) 1. 6:10 P66 D G pc it 1. (AW Byz TR) or omit

(P75 B S L H ) 6:40 6:53 6:57 P66 D pc b sy8 P66 D a om.

I i

1. .

66 D pc 1.

96
6:62 7:12 7:14 7:42
6:3 6:5

66 D

1. om.

66* > e syc

66* D 565 1. 66 D U 124 157

66 D

om.
1.

66* D it

1.

6:66
7:1 7:13 7:17 7:29 7:31

66 K D e t p p c b c f l
66 D 248 314 892 it 66 D 33 it vg syCP P66 D omit

add
omit xat .

P66 D 131 1. 66 D 6

1. ex 6 7:32 7:32 66 D pc c e P66 D add 6 . ot

1. ot xat ot 7:37 7:37 7:40 P66* vid D 1 22 69 pc it vg 66* S * D b e om. 1.

66* D (it) 1. (TO )

7:46

66* * D

(Byz TO ) Although most of these are the less significant of the Western readings in these two chapters (one-third of them involve word order alone), they are of such quantity

97
that there can be little question that P66 here has a strong mixture of Western readings.
1

However, along with this influx of Western readings

into its text, P66 continues to have a steady mixture of the "Byzantine type" of reading:2* (Again, only those

readings are cited where P66 differs from the Neutral tradition, usually as represented by P75 B, and is supported by a majority of later MSS.) 6:2 eOewpouv P66c P75 A D L f 053 pc e D o v P66* Byz TR opo y 75 R D L 579 P66 Byz TR 75 66 ( t A C D V Byz TR C W Db c e P66 St A L Byz i t p l TR

6:10 6:42 6:49 6:55 6:57

1

ev ev epnut 0 inoet ii|

P66 c P75 C W L pre 66* K* D Byz TR P75BKCLTHKpc 66 W Byz TR D b q

6:58 6:71 7:3

inset P75 B S C V L T p Q 66 D Byz TR 75 B C D V L H 6 9 124 pc 66 Byz TR 75 (*) D L pc 66 Byz TR

*Some of the "Western" readings given above are a l s o of t h i s t y p e . Cf., e . g . , t h e f i v e i n s t a n c e s o f the a d d i t i o n o f a conjunction ( 6 : 7 , 1 0 , 6 6 ; 7 : 3 2 , 5 0 ) .

98
7*12 7:16 a U o t 6c 7 5 I t P l IS P66 D L Byz pier

33 66 D L W Byz TR 7:29 75 L Byz 6e P66 D 33 pm TR 7:39 P75BEKKSUVAPC

66 K D V L Byz .

7:39 omit P66C 75 i 42 91 arm t 66* W L Byz 51 oytov 053 254 e sy^l11 * e r D f r 7:41 ot 6e P66c P 7 5 B W L N T X 8 3 3 X 5 6 5 ItPl p66* D Byz r* sy 6c 71 485 543 P TR 7:42 epxerat 75 W L 33 1071 c vg P66 D X Byz ItPl TR As in chapters 1-5 the majority of these are of the less important type of variant, and they again are chiefly secondary to the readings of P75 B. This strain of read ings, therefore, continues much the same in these two chapters as in chapters 1-5 and, as will be shown later, continues with varying degrees of frequency throughout the Gospel. These two lists clearly show that the basic reason for the sudden shift in the percentage of agreenents which P66 shows in John 6 is the result of mixture froa the Western textual tradition. (Note how often even in the second list P66 agrees with S or D or both against the Neutral MSS.) In order to gain a clear perspective both of the full extent of this mixture, as well as of the nature

99
of the relationships between 266 and individual KSS, a comparable list of 266 and/or D singular agreements for the remaining chapters of John is helpful: 8:23 ?66 * 1574 1. 8:24 266 140 244 348 pc a e . 9:9 266 it oa. ort

11:14 266 V* oa. ante 11:35 P66* K oa. ante 12:18 266 R airrov 1. 14:15 266 33 PC 1. or 16:15 266 oait verse 16:24 266 V* 1. 17:12 266* * sy3 oa. () pot 17:23 266 1 1. (or tva) 19:3 ?66 1. ot .

19:15 266*

8:35 266 070 pc a ff2 vg add 8:39 P66 D e add 8:42 266 D c e f ff2 1 q . 8:44 266 D 053 1. 8:48 266 D L 0124 pc aur c r* 1. 8:53 266* D a 1. 9:10 266 D b rl bo 1.

100 9:18 P66 D X pc . 9:19 66 D 1. 9:27 66 D a e r 1 -naXtv 1. 9:28 6 D 053 PC a b c ff2 vg 1. tt 9:38 P66 D It et

1. 10:4 66 D 124 700 b c ff2 q 1. 10:10 66 D ff2 om. 10:16 266 D add 11:7 66 D P45 435
e

fa 1.

11:28 66 D W add 11:30 66 D 1. 11:32 66 D 579 pc q 1 om. 11:33 66 D 45 11:37 66 D it 1. 1. 1. xat

11:45 66 D 45 a b f r* 11:51 P66 D om. 11:52 D P45 12:2 12:3 (P66) D

om. ante

P66* D 1194 b c e ff2 rl om. . .

12:19 P66* D 579 1241 a c e

12:26 P66 D W 1093 pc a b c e rl 12:30 66 D It 12:31 66 D 1.

1.

101 12:31 P66* D W b ff2 1 rl vg om. 12:40 6* D a e f 1 add ante 13:10 P66 D 235 sys add 13:34 P66 (D) pc it add 15:7 15:8 P66 D e om.

P66 D 254 .

15:17 P66* D e om. tvo 15:24 D it om. xat

1

16:21 P66 D 248 it syS 17:10 P66*vld D

1. cupa

17:14 P66* D * 69 om. . . 18:9 P66 D 42 122 1.

10:25 6 D

om. t 1.

12:32 D 56 it vg

15:13 D* a b c e ff2 om. 18:10 D 242 it 1.

At least two important conclusions may be reached from the foregoing sets of lists: 1. In contrast to chapters 1-5, a much greater part of the variation in from the Neutral tradition in chap ters 6-20 is in favor of a mixture of Western readings. These are more abundant in chapters 6-7, but remain in varying degrees throughout the rest of the Gospel (espe cially in chapters 11-12). A sufficient number of these

102 readings are of such significance as to indicate that a mixture of patently "Western" readings is involved (cf. 6:64; 7:40; 11:45; 12:30, 40; 13:10; 16:21). 2. The nature of the relationship between P66 and It is probably safe

is more clearly brought into focus.

to say .that a significant relationship between these two MSS exists only in John 6-7. The only important unique

agreement they have after these chapters is the common omission of 16:15. But rather than to explain this in

terms of textual relatedness, as does Porter,25 it seems more likely that this is an independent case of homoioteleuton, the kind of error to which the scribes of both MSS are commonly addicted. Furthermore, the relationship which does exist between P66 and in John 6-7 is to be explained in terms of a "mixture" of Western readings in P66, in a section where is already a decidedly Western text.2** The lack of

5 " A Textual Analysis," p. 150. Apart from his list of "agreements against the rest" (see supra, pp. 32-33), this is the one reading he singles out in the entire Gospel as an argument for relatedness. Porter ( A Textual Analysis," pp. 149-150) ob " served the increase of agreements between P66 and in chap ter 7* but as to the reasons for this phenomenon, he con cluded: "There is not enough evidence available to reach a decisive conclusion," (p. 150). This hesitation is proba bly the result of his methodology. The data seem clearly in favor of his first alternative: "It might be taken as an indication that cither the text of P66 or the text of Codex Sinaiticus changes in chapter seven." It is the text of P66 which changes. . .
26

103 any close relatedness beyond this point is due to the fact that, in spite of a continuing mixture of Western readings in P66, is no longer Western.27 Something should be said at this point about Boismard's hypothesis that the mixture in P66 is the result of the scribe's having followed two Vorlagen in clearly defined sections. Although the MS as it now stands appears to indicate that two Vorlagen were used, the second one probably was not used for the original, but for corrections only.28 Hiere are at least two reasons for finding a hypoth esis other than Boismard's for the mixture in P66. In the first place, Boisraard himself qualifies his hypothesis by noting that "meme quand il adopte l'un des textes, [6] garde des traces du texte concurrent, dans une proportion plus ou moins forte."29 Either this is the case, or else

the two Vorlagen were themselves mixed texts; for clearlydefined sections of Western text simply cannot be found. In the second place Boismard is correct in looking I 'In chapters 9-21, where P66 and are both basic ally Neutral with varying degrees of mixture, each has a closer relationship to than either has with the other. In the numerous instances where either MS varies from B, each does so more often without the support of the other, than they do together. ^ S e e infra, pp. 16? ff. for the discussion of the corrections.
2 2

9"Le Papyrus Bodmer II," p. 389, n. 1.

104 for evidences of conflation, and one should expect these, if his hypothesis were valid. However, his list of ten such readings is less than convincing. Three of them are

the result of errors in the editio princeps (4:15; 6:58; 11:2). In 4:15 the original readsfcttnowith D, not The m of the present text is clearly

btfwom (as Martin).

secondary, and traces of the are still to be seen under it. The reading in 6:58 was simply a transcriptional error on the part of the editor. In 11:2 the original text of

P66 reads xat . Even if this is a contamination of the "Prater eius erat ille Lazarus qui infirmabatur,, of the Liege KS of the Diatessaron and the Slnaitlc Syriac, the present text is not a conflation of two MS traditions, but a correction fro one reading to another. The rest of Boismard's "conflate* readings are equally dubious. One vonder3, for example, how "Western" is the addition of in 6:6l, which is supported only by Nonnus and one MS of the Ethiopic version, or the ou of 10:28, where D is supported by K, L, and X. The double ouoevt in 8:33 may be explained as a conflation of two MS traditions, but how does one explain the similar repetition of (without correction) by A in 2:13 or of the oux exere by fc* in 5:42? Furthermore, it should be noted that where the scribe does create a conflate reading in his corrections (14:14 C Y ) , the one reading (ere)

105 is not strictly Western, where It is read by V Q and the entire Byzantine tradition, and the other reading () is supported by A and c q r* vg as well as by P75 L et al. Although his work is helpful in many ways, Boismard's major hypothesis seess to fall short of clear demonstration. 3he change in P66 is probably to be accounted for in terns of a mixture which existed already in the Vorlage from which the scribe was caking the origi nal copy. Furthermore, except for chapters 6-7* the greater amount of mixture in F66 is not from readings of the western tradition. As will be shown in the analysis of the remaining chapters of John, P66 retains a higher mixture of Western readings than appears in chapters 1-5, but the increase of "Byzantine-type" readings is a such more significant factor. John 8-9. Table V (pp. 86-87) indicates that in these two chapters P66 has no close relationship to any single MS, although it is closer to the Neutral HSS than it is to D. An examination of selected readings suggests that the basic pattern of relationships is similar to that of chapters 1-5, not 6-7. The chief difference is the in crease in the number of variants where P66 reads alone (or almost so) with J> among the Greeks (twelve instances; see the list on pages 99-IOO). Besides these and the two agreements with (8:23,24), there are two other readings

106 which may r e f l e c t the Western t r a d i t i o n : 3 9:18 9:27 P66* 565 660 I t bo P66 22 aur b e e oa. 0* ff2 1 o e . oux

I t Is a t once c l e a r , however, t h a t the r e l a t i o n s h i p of P66 t o the t e x t l a t e r found in D i s s l i g h t indeed In comparison with chapters 6-7.31 Even In the sections where Bolsmard considers P66 t o be following a Western Vorlage (9:l8b-19 and 9:26b-39), the r e l a t i o n s h i p i s so s l i g h t t h a t

30Boisaard, "Le Papyrus Bodaer I I , " pp. 366-387, a l s o Includes as "Western' the omission of in 9:30, as well as the 1. ev ? and the unique word order xat etnev 1. xat etnev. But the methodology here seems open to question. Bolsmard recon s t r u c t s the "Western" reading from four separate readings variously found in D, Nonnus, Tat Ian, and some of the ver s i o n s , but supported 1In t o t o by none of them. The "agreement ' of 6 and D in omitting t seems doubtful, since P66 does so in an a l t e r e d word order. I t would seem a l s o t h a t the reconstruction of a "Western" reading so scrambled as the one Boismard presents must take place a t a more s i g n i f i c a n t point than a t one of the Johannine formulae for introducing d i r e c t discourse. Cf. K l l j n , "Papyrus Bodmer I I , " pp. 329-330, where he argues for P66 D relatedness from a s i m i l a r v a r i a t i o n in 10:3^. Klijn here has misread D, and B i r d s a l l , The Bcd^er Papyrus, p . 15, has c o r r e c t l y offered t h a t "we have here . . . independent attempts a t a s s i m i l a t i o n " (although 3 i r d s a l l , t o o , has i n c o r r e c t l y c i t e d D). Relatedness a t a point l i k e t h i s must be found in KSS which c l e a r l y a g r e e , such as the unusual (for John) formulation, () , of P75 (L 157 1321) a t 12:30.
Of the twelve "singular" agreements, five involve word order, two add a conjunction, one involves a compound verb form, two are shorter forms of similar expressions, one replaces an imperfect with an aorist, and one involves a correction in P66 of the kind which indicates a scribal error rather than a change to another textual tradition. These are scarcely the kind of readings on which textual relationships may be built.
J

107 he must use Tatian, the Fathers, and various readings from the versions to find readings to support his theory. But however slight this relationship may be, it is greater here than in chapters 1-5. The fact that is still basically Neutral may be seen in the number of readings it shares with this tradi tion against most of the rest: 8:19 O Y i\beive 75 P39 W L 33 pc itP1 Byz TR 0 \ noetre D 209 b e ffr \ 8:23 75 W 13 124 1010 1293 D L Byz TR 8:25 etnev 75 W L 69 124 pc ctnev D 249 892 1241 xat ctncv Byz TR

66* 75 476 P66 c W L Byz TR 8:28 omit * W L 1 55 1241 P66 c 75 D Byz TR 8:28 75 sa bo omit K D V L Byz TR 8:34 (75) 0141
K C D W L N X e Byz TR

8:25

8:38 75 C W 565 bo ertD D L pc era o Byz TR 8:38 P75 C W L 213 rl 1 vg D Byz TR 8:38 20 8:38 75 C W L 55 PC D Byz TR

omit 75 W L > C D Byz TR

108 8:52 94 einav P66 R C W 579 a b e r l ctnav 75 D L Byz itPl TR . . . P66 P75 * W L sa . . . 0124 bo . . . C Byz TR . . . . . . . . . P66 P75 L 33 565 579 D C W Byz TR

9:6

9:9

9:9 9:11

66 75 C W L 33 1241 pc 124 Byz TR D , 66 75 C W L () 33 pc OTt A D Byz TR 66 33 P75 vid C 565 579 A D W Byz TR 66 75 i W L 33 a b c ff2 t A D Byz TR itPl omit P66 P75 W L X 33 pc A D Byz TR 66 P75 6 F pm omit D W L G U X 33 pc TR 66 75 W A L X Byz TR D 66 75 W L 157 124l pc A D Byz TR 66 75 L 33 1241 A D W Byz TR , , , , . 66 75 A D W L Byz TR 66* 75 W P66c D Byz A L pc TR .

9:14 9:20 9:20

9:23

9:24 9:30 9:36

9:41

109 It will be observed that, in contrast to its agree ments with D, P66 here shows a considerable number of textually significant readings. As in chapters 1-5, the chief cause of "mixture" in John 8-9 is the tendency in P66 toward a "Byzantine-type" of reading. I note the following: 8:l4 75 W 39 0l4l pc -rtv 66 Byz TR D 75 D W L 33 213 892 1241 66 Byz TR Gixata 544 75 C W L 33 pc 6 * D 070 Byz TR W L 070 1 713 1321 bo it ctnav P66 P75 C D Byz TR P75 C D W L 33 258 1241 6 Byz TR P66c P75 * C D W 713 66* A L Byz TR xav P75 D W 1170 xat 66 A C L Byz TR

8:16

8:38 8:41 8:51 8:54 | 8:55 j

8:55 75 A D W 565 157 52 254 66 K C L Byz TR ! 8:58 9:9 9:12 75 C 579 66 A D W L Byz TR 75 C D W L Byz TR 66 * 33 pm it xat 75 W L 33 565 pc 66 D Byz TR A e sa bo 75 s D W L U 33 66 A N Byz TR

9:19

110 9:26
t

omit P75 D W itP1 Po6 Byz f q TR * W sa TR c 33 565 PC X Byz pier

9:28 xat 75 eXot6opnaav 69 pc ot 6e D L 66 A 9:35 omit P75 D W e axxrw P66 A L Byz TR

Again, the readings of P66 appear to be generally secondary to those of P75 et al. While these are not the kinds of readings whereby one establishes textual relationships,

they do indicate the tendencies in P66 toward a smoother and fuller text. John 10-14. What has been noted of P66 in John 1-5 and 8-9 increases in chapters 10-14. The amount of mixture

from the "Western" tradition is slightly higher in 10-12 than elsewhere (except 6-7), but the largest number of variations from the Neutral MSS are the result of "Byzan tine-type" readings. There are, for example, in these chapters sixteen instances in which P66 adds a conjunction to remove asyndeton (10:12, 16, 19, 21, 31, 32, 39; 11:22, 32; 13:2, 22, 23, 25, 26, 28; 14:7), as compared to two

^This is true even of 9:28, where one would usually argue that asyndeton best explains the divergence of conJunctions. Such seems not to be the case here. The full reading of P75 et al. is xat xat etnev. One would have difficulty explaining how a scribe should have preferred the paratactic . . . xat to any of the other alternatives.

Ill such instances in chapters 1-5.33 There is also, again in contrast to chapters 1-5, an increase in the number of more significant variations from the Neutral MSS. In each of the following variants P66 Is now the earliest witness to the reading of the Byzantine tradition: 10:26 11:29 omit eiTTov uutv . . . . . . . , . . . . P66 c P75 V L pa c vg P66 A D X Byz aur a b e 051 75 B K C W L X 3 3 a b P66 P45 Bvz TR D aur c e f r l

11:31 11:32 11:57 12:1 12:6 12:22

(75) B ' K C D V L X M 700 pc 66 Byz i t vg sa I S P75 S C D L 33 579 892 6 Byz TR

V 065 28 565 579 PC 66 A D L 8 Byz i t vg TR V L 213 a c e r 1 sa syP 66 8 065 Byz aur b f f f 2 1 TR P75 D V L Q 33 565 579 892 vg P66 65 X Byz i t TR

omit , e t x e v , xat

13:2

. . . epxerat . . . xat P75 A L a xat P66* D W Byz i t P 1 TR xat epxerat . . xat 157 omit P66 c c sa V L 579 124l pc d 66 A S 9 Byz itP 1 vg TR

33rpne figures at this point include only those instances where P66 is Joined by at least one other important MS and usually by the Byzantine tradition. Excluded are singular readings and corrected readings.

112 13:12 * C* 157 579 bo a e 1 66 A L 33 1241 pc ItP van co t D Byz vg TR o 13:18 C L 33 892 1241 pc P66 A D V 8 Byz TR 13:18 C L 127 213 249 892 1071 1093 ?66fitA D V Byz It TR 13:26 Jepjeata . . . trow () . , . entbvcw bwaw D L 213 713 2241 sa bo P66 A D 8 I Byz TR V

13:26 C L 33 213 1071 1241 1321 omit P66SAD8Byzit?g2S 14:4 14:5 P66c C V L Q I 33 157 a P66* A D 8 Byz ltP^ ^ TR

C* (D) a b e bwaixeQa P66 () A V L Q. Byz TR 14:14 75 A L 33 124 pc c 2rl P66* S D V Q Byz aur a d f ff TR P66c 1241 14:16 75 L Q 33 It 66 A D W Byz TR Again, this list does not mean that P66 Is tbe fore runner of the Byzantine tradition. It shares xaany of these readings with the OL, and the TR here say simply reflect the Western tradition. But the chief characteristic of the majority of them is that they are secondary to the Seutral reading. Even von Soden, who includes csore "Byzantine" readings than most of the critical editions, reads with P66 here only at 11:32 and 14:5. (He adopts the conflate reading at 12:22.) John 15-21. The nature of the textual relationships of P66 in this section is more difficult to determine. Not

113 only Is P66 fragmentary, but P75 is totally lacking, and C, W, and D suffer major lacunae. However, the list of agree ments with J> on pages 99-101 Indicates that the mixture of Western readings appears to be in about the same proportion as In chapters 8-14. The amount of variation in favor of

"Byzantine-type" readings appears to be much like that of chapter 9 A direct examination of P66 in chapter 19 indicates that P66 is still basically Neutral, and within that tradi tion it is more closely related to than to K. This is made clear from the following lists of readings: (An * in

front of the verse indicates that shares the reading of the T R . ) ^ 1. Variants where P66 agrees with against the Byzantine tradition. 19:3 19:3 19:7 19:7 19:11 19:11 oa. 1. . post 1. 1. 1.

^ is lacking in this chapter, but the amount of agreement between P6b and D on each side of the lacuna is much the same, so that one may assume it would be here as well. The advantage of using this chapter for the analysis is that the relationship of P66 to and may be clearly brought into focus.

114 19:12 1. 19:13 . 19*17 1.

19:20 , , 1. , , PuipatoTt 19:29 1. , 19:31 cnet . . . , . . . 1. . . . , en . . . 19:3^ 1.

19:35 * 1. 19:35 1. 19:40 1. 19:41 1. 2. Variants where P66 agrees with members of the Byzantine tradition against : 19:17 1. 19:35 1. 19:39 1. 19:39 1. * 3. Variants where P66 agrees with against K: 19:1 19:4 19:4 *19:6 . . . 1. xat 1. (TR )

1. om. post

115 19:10 19:12 19:23 19:27 19:28 19:28 19:28 19:29 19*33 19:35 19:38 19:38 19:38 19:39 *19:39 19:^0 4. 19:3 19:4 19:7 19:15 1 . 1. (TR )

ore 1 . ot 1. 1 . 1. 1. 1. add (TO add )

1 . *at 1. 1 . . . . 1. . . . 1 . 1 . 1 . 1 . Variants where ? reads w i t h a g a i n s t B: 1 . 1 . om. 6 1. (TR ot ) (TR )

19:16 *19:21 *19:31

1. omit 1. 1.

116
III.
i

CONCIUSIONS

Prom the foregoing analysis one may make the following conclusions about the textual relationships of P66: 1. P66 is a basically Neutral text. 2. ?66 has a closer relationship to P75 and than

to the other MSS of this tradition. 3. P66 as a whole is closer to this tradition in

John 1-5 than thereafter. 4. The greatest amount of variation from this

tradition in John 1-5 has tendencies toward a "Byzantine type" of reading. 5 Prom chapter 6 on, there is an increase in the

amount of mixture from the Western tradition, the greatest amount of this mixture appearing in chapters 6-7. 6. The greatest amount of variation in chapters

8-21 is the result of an increase of "Byzantine-type" readings. This is the chief reason for its less close

relationship to the Neutral tradition in these chapters. 7The alleged close relationship between P66 and

exists only in John 6-7 and is the result of agreement in readings within the Western tradition. When becomes a basically Neutral text, in chapters 8:39 ff* P66 is more closely related to P75 and than to K. 8. Although a direct relationship between P66 and

 doe8 not appear to exist In John 8-21, the observation of Zimraermann seems to be valid for this section of John, namely that P66 shows substantially the same characteris tics which constitute the peculiarities of (supra, p. 78) In view of these conclusions, it seems correct to refer to ?66 as a mixed text. It is basically a witness to the Neutral tradition found in its contemporary P75. But it varies from this tradition in a profusion of readings from the Western tradition and other, chiefly secondary, readings of the type found later in the Byzantine tradition. Contrary to Klijn's critics, it seems to this writer that his description of ?66 as "Neutral in a non-pure way" is altogether fitting. j

i i
i

(
I I

CHAPTER IV TEXTUAL AND SCRIBAL CHARACTERISTICS OP P66 One of the most important questions in the search for the "original" New Testament text is to determine the nature and amount of editorial activity in the KS tradition, while it is true that a certain nuaber of variants have come into the tradition as "errors3 on the part of scribes, and therefore are the result of 'non-editorial* activity, it is also true that a large part of the variation is the result of scribes1 choosing to add, delete, or alter cer tain words. How many variations in a given MS can be attributed to either of these processes and how many are already in the scribes* exemplars is not at all easy to determine. But since most important variations are proba bly the result of "editorial3 activity, the search for such activity in the earlier MSS is of great importance. It is here that 66 may prove to have its greatest value. Its importance at this point Is the result of two factors. In the first place, if our analysis in Chapter

III is correct, P66 Is an early witness to a distinct 118

119 textual tradition (the Neutral), but It is so as a 'cor rupted" or "mixed* member of that tradition. It is there fore possible to indicate in some measure the cature of its text by indicating the general characteristics of its variation from its basic tradition. Secondly, P66 is the earliest New Testasent KS In existence in which editorial activity in the form of aoajor corrections to its text has taken place. Since most of the corrections are probably by the original scribe, and aany from another than the original Yorlage, one can determine to some extent the nature of the "editorial* activity. It Is in the pursuit of these textual and editorial characteristics in to which this present chapter is devoted. The results of our analysis, If valid, should have at least three important consequences for KT textual criticism: (l) It should enable one to evaluate the gen

eral worth of 66 as a whole in the search for the "origi nal" NT text. (2) It should further enable one to evaluate (3) It

P66 as "external evidence" for any given reading.

should give information as to at least one kind of recensional activity which existed (presumably) in Egypt at a very early period in the transmission of the Kew Testaaent.

~These two conclusions about the corrections of P66 have been set forth by the present writer in "Early Textual Transmission." Cf. the discussion infra, pp. 167 ff.

120
I. TEXTUAL TENDENCIES OP P66 WHERE IT DISAGREES WITH ITS BASIC TEXTUAL TRADITION In the preceding chapter It was noted that J. N. Birdsall broke new ground in his analysis of P66 by attempt ing to assess its nearness to the original NT text by means of grammatical and stylistic criteria (supra, p. 8l). The

present study proceeds along the lines he laid out, with one significant difference. In the present study the

attempt has been made to assess P66 both in terms of its relationship to its own textual tradition as well as to the Johannine original. Therefore, instead of cataloging certain Johannine usages and examining P66 as a whole in their light (per Birdsall), I have chosen to examine priearily those readings in P66 where it varies from its basic tradition. These readings have been classified and exam

ined according to their frequency, with an eye for the tendencies exhibited at these points of variation. Some words of caution are necessary here. In the first place it is not always possible to ascertain the Neutral tradition. For the most part P75 agreement has been considered the safest clue. Where these two divide, or where P75 is missing, the tendency has been chiefly to

This means that P66 will almost always be seen where it differs from the Neutral tradition. It should always be borne in mind, however, that P66 is basically Neutral and more often shares readings with the MSS of this tradition than against it.

121 J
follow one or the other where it ie supported by two or more of the witnesses to this tradition (usually C L W; eooetiaes 33 579). It is acknowledged that this often

involves guesses. A study of the Neutral texttype along the lines laid out by Colwell is needed. In the absence

of such a study, and for the purposes of the present paper, the procedure followed seems adequate. In the second place, the decisions made here as to whether P75 or P66, where they differ, most likely pre serves the original text of John may often be open to question. For the most part the general canons of internal criticism have been followed, asking usually which reading best explains the existence of the other. Often this

decision is brought before the court of "Johannine usage." But a real difficulty sometimes emerges here. Is that read

ing to be considered original which best fits Johannine

3"The significance of Grouping of New Testament Manuscripts," NTS, IV (1957/58), 91-92. The position taken in this study is based on the rather common assumption that is the leading representative of this tradition. One could argue, of course, that the so-called "secondary" wit nesses to this texttype are its "best" representatives. However, a collation of these MSS with shows that most of their disagreements with are usually in conformation to the Byzantine texttype. Moreover, an analysis of the dis agreements on internal principles shows that the other MSS have the majority of secondary readings. If the Neutral texttype, as Zuntz (The Text of the Epistles, pp. 271 ff.) has argued and as is usually assumed, reflects 'Alexandrian philological know-how," then P75 and are easily the best representatives of this texttype. P75 agreement as the first clue to the Neutral tradition seems also to be verified by the analysis in Chapter V of this study.

122 usage, because the reading ls_ Johannine, or is that reading to be considered original which deviates from the Johannine norm on the basis of the canon, ardua lectio potior? In

some instances where Johannine usage seems fixed, variation is simply too difficult to be original (e.g., the singular word order uptv Xeyuu of at 10:1, 7; 13:21, or the single of at 3:5, in the apparently "fixed" formula []). But in instances where Johannine

usage is not so "fixed" two general tendencies of Judgment have prevailed: (l) Where Johannine usage differs from

more common Greek usage, that reading is probably original which is most like Johannine usage, on the basis that at such a point scribal "editing" would tend toward the more common usage. (2) Where Johannine usage is not necessarily

"un-Greek," that reading is often to be preferred which is less like Johannine usage, on the basis that at such a point it is the more difficult reading. With these cautions and guidelines in view, the following analysis is offered; it is based on 376 readings in P66 where it varies from the Neutral tradition. This

includes 98 readings where P66 reads alone (or nearly so) among the Greek MSS; but it excludes such items as orthog raphy, itacism, and obvious errors. Also excluded are the 217 readings which have been corrected and which will receive special attention in a following section. Word Order. The single most frequent cause of

variation in P66 from its basic tradition is the transposi tion of words. There are at least 78 such occurrences in the MS, 22 of which are singular to P66. Although there are seven such singular readings which have been corrected, the large number of these readings, both singular and otherwise, seems to indicate that the scribe tended to show a general lack of concern for the order of words. To be sure, many of the 56 transpositions which P66 shares with at least one other major MS are supported only by MSS of the Western tradition. But for the most part there seems to be no clear pattern of influence of one textual tradi tion on another. What may be more significant is that

Nestle-Aland (25th edition) reads with P66 against the Neutral at only four of these 78 places (8:14; 11:44; 12:30;

14:20).5 Since Greek word order is very free, it is not always easy to determine at any point which reading is

I I

more likely to be original." But an examination of some of ': 66 shares 28 of 46 transpositions with D; 23 of 56 with K; and 21 of 56 with the TR. ->This of course may only mean that a text like Nestle-Aland is under the dominance of the Neutral tradi tion. The latest study of Greek word order is that by Kenneth J. Dover, Greek Word Order (Cambridge, i960). This work deals only with classical Greek. A study of this type for the koine period would be helpful. Without such a study, one must rely on what is available in the Grammars and on what may be determined to be Johannine usage.

124 the transpositions in places where Johannine style may be determined indicates that for the most part P66 appears to be picking up secondary readings. 1. Position of subject and verb. Thirteen of the 78 instances are transpositions of the subject and verb, and in five of these, the nominative personal pronoun i3 involved. Nigel Turner has observed that in the NT personal pronouns are often inserted where they would be unnecessary in classical Greek. This is particularly true of the

nominative, which in classical Greek was usually not em ployed except for emphasis or antithesis. This principle is not strictly observed in the NT, particularly not in John, who uses the nominative personal pronouns more fre quently than all the Synoptists together.** Where the nominative pronoun does occur in John, it almost invariably precedes the verb. This is especially true of eyu) (approximately 129 times), except for the five occurrences of the phrase eipt (7:34, 36; 12:26; 14:3; 17:24). In 12:26, P66 and D read eyw and P66

^Syntax, Vol. Ill of A Grammar of New Testament Greek by James Hope Moulton (Edinburgh, 19*>3)> 3* 37-38. A great part of John's usage does involve antithe sis, but there also seems to be a free use of the pronouns where neither emphasis nor antithesis may be observed. Cf. the discussion in Edwin A. Abbott, Johannine Grammar (London, 1906), pp. 295-298.

125 does eo singularly In 17:24. P66 Is here conforming to the far more common etpt, and Is clearly secondary. With the pronouns , , and there are only eleven out of some 141 possible Instances where the Nestle-Aland text reads the personal pronoun following the verb. At four of these the major KSS show textual varia tion (6:30 ; 8:48 ; 12:34 ; 14:20 ). The seven occurrences without variation indicate that this order is not a scribal error, but an occasional Johannine usage. Of the four readings with variation, P66 has a lacuna at 6:30, but reads against the Neutral tradition in the remaining three: with the pronoun first at 8:48 and 12:34; with the pronoun second at 14:20. Except for 14:20, therefore, where 66 perhaps reads the original order, there seems to be a tendency in the KS to harmonize with more prevalent usage. Such harmonization is probably a secondary procedure. On internal grounds alone one cannot decide in other instances where P66 has a transposition of subject and verb, except to note that there is a tendency to read the subject before the verb (5:18; 7:42; 8:44; 11:21, 30; 19:1; only 7s35 reads verb-subject) .9 Furthermore, there is one

^Cf. the discussion in Turner, Syntax, p. 347, n. 2, where he notes that "the normal order in the ancient Greek was SubJ.-ObJ.-Verb (SOV). . . . Some NT books approach

126 of these which seems clearly to be secondary. At 5:l8 P66 and D read ot Iouoatot for the ot anoxrctvat of most of the rest If the reading of P66 D were original, it would be diffi cult to explain the unusual order (verb-object-subject) of 175 re 11. 2. Johannine "Variation." Two others of the

subject-verb transpositions are involved in another marked characteristic of Johannine stylehis fondress for "varia tion." This has been described by Abbott as *the habit of repeating the same thing (or representing his various characters as repeating the same thing) in slightly dis similar words and with slight dissimilarities of order."1 Although this "variation" may take several forms, it usually involves a change of word order such as AB BA or ABC CBA. A good example of this characteristic, both in word order and in slightly different words, is 1:48-50, eibov . . . ct6ov * . Abbott has included a considerable list of instances of such "variation" in John. Although his list is not

this standard, but on the whole NT is closer to the Hebraic order (VSO) and towards the subsequent tendency of odem Gr eek (SV0).n The tendency here in P66 to read the order SVO appears to be a step in the direction of this "subse quent tendency" of Modern Greek. Grammar, p. 401.

127 necessarily complete, it serves as a convenient guide against which one may check the tendencies of P66. An examination of the entire list reveals that the majority occur without textual variation, thereby indicating that this is a true characteristic of the Fourth Gospel. Of our 76 transpositions, there are seven which are found on Abbott's list. In six of these P66 has a reading which tends to eliminate this feature of Johannine style: (The underlined portion indicates the point of "variation.") 1:49 P75 266 et utoc , et . et , . . . . . . . . . . .

6:31* *^9 75

66

! 74-2 75 ; OTt . . ; 66 ; . . . ; 8:14 (cf. 531-2) 75 . * . . . , . 66 . c * . . . , .

128 9:28 75 P66 11:29-31 et exctvou. coucv \\. exctvou ct. 6e .

P75 () () . . . OTt xat . 66 () xat () . . Maptav xat .

At still another point P66 keeps the variation, but reads the exact opposite of the Neutral order: 8:51-2 P75 eav euov , . . . xat uou . . . 6 eav euov , . . . . . . There is one other such variation, not included in Abbott's list, where has a reading which seems to violate this feature of Johannine style: 15:9 75 66* , . . .

It will be observed that in most of these, the variant reading in P66 occurs in the second portion of John's "variation.11 This seems to be a good indication

that secondary processes of harmonization are at work.*-1

^Another set of variants on this point occurs in the three occasions in John 7 where the people are speaking or murmuring "concerning Him" (7:12, 13, 32): P75 nept . . . nept . . . . . nept 66 nv nept . . . nept . . . nept

129 1 3 The "vernacular possessive." Another character

istic of Johannine word order is what Abbott has called "the unemphatlc precedent possessive , or 'the ver nacular possessive.1"*2 This means that there is a tend

ency in John for the possessive pronoun to precede the noun as against the more common Septuagintal form where it usually follows. John thus uses the possessive twice as often as all the Synoptists together. In our list of 78 transpositions, six involve the possessive pronoun. At five places (1:27; 6:53; 9:21; 12:16; 18:10) P66 reads against the Neutral tradition in adopting the more common order of placing the pronoun after the noun. Only at 10:4 does it read the "vernacular pos sessive" against the Neutral tradition, where it is joined by D 124 700 b c ff2 1 q. It would appear, therefore, that P66 at this point has a tendency away from Johannine style and toward a more common form of expression. Again, this probably indicates a secondary process. These readings may have no relationship one to the other, and probably should be judged individually (which judgment is suspended here because there seems to be no criterion by which to Judge on internal grounds alone). But it is curious that textual variation should occur at each Instance of this similar expression, and that each set of readings should tend to keep a form of "Johannine variation." ^Grammar, p. 4l6.

130 ' 4. Partitive Phrases. In Abbott's discussion of partitive phrases in John, he notes one characteristically Johannine feature which involves word order, namely, the use of the partitive genitive, with or without -ex, before the governing word.*3 Abbott lists twelve such occurrences in John. At two of these P66, with other MSS, inverts the word order to a more common form: 6:64 7:31 a\\a ctotv ot
ex

P75 C W pier 66 S pc

75 L pc 66 D Byz TR

The same thing occurs in P66 at 7:40, but has been correc ted to conform to the Neutral tradition. P66 here is clearly picking up secondary readings. 5 There are a number of transpositions where the reading of P66 appears to bring the elements of a sentence into more logical Juxtaposition.!^ All of these are

13Abbott, Grammar, pp. 89-90. Cf. Eduard Schweizer, Ego Eimi . . . Die religlonsgeschichtliche Herkunft und theologische Bedeutung der johannelschen Bildreden, zugleich ein Beitrag zur Quellenfrage des vierten "Evangeliums_ (GSttingen, 1939)* P 92. *^What one considers "logical," of course, may only be a value-judgment; for word order in Greek indicates emphasis. However, emphasis sometimes appears to separate elements which normally go together. Cf. F. Blass and A. Debrunner, A Greek Grammar of the New Testament, 10th German ed., trans, and rev. Robert W. Funk (Chicago, I96I), par. 473. Where such separated elements are brought closer together in variant readings, one may assume that the read ing where such elements are separated gave rise to the other(s) and is therefore more likely to be original.

131 probably secondary, as they are less likely to have given rise to the alternative reading: 1:28 9:38 PJ5 75 66 cv vat ev atTiav aiTtav

11:17 75 66 Byz TR 12:18 17:5 75 66 C 66

18:38 L 66 Byz TR 19:35

L 66 By the same criterion, however, there are some

readings where P66 may preserve the original. This is especially true of 19:4 (cf. 18:38 above): 1 33 73 138 66 W L pc atTiav psupp g ^ 35* * There seems no valid reason to choose the reading of here, except on the basis of "Johannine variation"as a variation from 18:38. But P66 also involves "variation" i

and seems much more difficult to derive from the reading of than vice versa. One may conclude, therefore, on the basis of an

132 examination of word order variants in P66 in light of Johannine style that P66 probably preserves soae original readings; but for the most part where it varies from its basic tradition, it tends to pick up secondary readings. Horeover, most of these secondary readings create an easier text or more common Creek style. P66 seldom varies from the Neutral tradition toward a more difficult text. Conjunctions. The second major cause of variation in P66 from the Neutral tradition is the addition/omission/ alteration of conjunctions. I note 50 such instances, ten of which are singular readings in P66. Of the 50, six only are omissions in 266,15 nine involve an alternative con-

junction,1^ and the remaining 35 involve the elimination of asyndeton in P66.1? Hie Johannine characteristic of asyndeton has often been noted.* Schweizer, in his discussion of Johannine

characteristica, tried to define the term more sharply, and isolated 39 cases of what he considered to be true asyndetic sentences in John. However, the amount of variation

15 l6 17

7 : 1 , 12; 8:24; 9:28; 13:26; 14:19. 4 : 4 2 ; 7:28; 8 : 2 3 , 2 5 ; 9:12, 2 8 ; 11:5*; 13:26, 2 9 .

1 : 1 7 ; 2:16; 3 : l 8 ; 4:30; 6 : 7 , 10, 66; 7:29, 32, 50; 8:33, 35, 39, 4 l ; 9 : 9 , 2 3 , 2 9 ; 10:12, 16, 19, 2 1 , 3 1 , 32, 39; 11:22, 32; 13:2, 22, 2 3 , 2 5 , 2 8 ; 14:7; 16:33; 18:31, 38. ^ A b b o t t , Grammar, pp. 69-73; Schweizer, Ego Elml, pp. 91-92; Blass-Debrunner-Punk, Greek Gra-rryir, p a r . 4o2.

133 In the MS tradition throughout John Indicates that it is much more extensive than this. Abbott has given the following description of Johannine style at this polntr1^ John abounds in instances of asyndeton of the noat varied and unexpected kind, too numerous to quote, especially with an initial verb . . . ; with any form of the pronoun "this*; with the conjunctions "if* and "even as n ; with an adverbial phrase . . . ; with a participle with the article. . . . There is hardly any part of speech, or word, that might not coae at the beginning of a Johannine sentence without a conjisction. He concludes his discussion with a long list of such instances with an attempt at classification. A check of P66 against Abbott's entire list indi cates that for the most part, the MS tends to keep Je>annine style.20 However, the 35 instances where P66 reads a conjunction against the asyndetic Neutral text further indicates that a secondary process is also at work.2* Moreover, an examination of the six instances where P66 reads without the conjunction indicates that here, too, it is usually secondary. The reading at 9:28 has already been Judged to be

^Grammar, p. 69. This may also be shown by contrasting P66 with A and another of the later 3yzantine MSS. Poo far more often reads with than against it in preserving asyndeton. Birdsall, The Bodmer Papyrus, p. 13. "In respect of the stylistic criterion of asyndeton we nay con clude then that the papyrus probably preserves a number of original readings, but also displays a marked tendency to smooth over certain harshnesses in the original text.

134
secondary (supra, p. 110, n. 32). Another instance (14:9) is singular to P66. Since the scribe probably copied by syllables,22 a singular reading such as this, which involves only a syllable, is suspect as a mere scribal lapse. Another instance (13:26) involves the elimination of o w in the formula [] . Since John's ordinary style at this point is to read without the

conjunction, the of C L I pc is perhaps more difficult to explain than its absence in P66 pier. The elimination

of xat in 7:1 is not so much a case of asyndeton as it is the elimination of a xat from the beginning of a sentence which clearly starts a new section in John. Here one should expect asyndeton, and the xat is the more difficult reading. At only one point is the decision more difficult. In 7:12, for the second division of the crowd, P75 W pc TR read and P66 D L pier read . The prob j

lem here is whether the absence of is secondary and to be explained as conforming to a Johannine habit when indi cating various divisions of crowds (cf. 7:^1; 9:9* 16; 10:20, 21; 12:29) or whether its addition is secondary and to be explained from the preceding , thereby improving the style. Probably the omission here is the more diffi cult reading and is to be considered original (WH put it in

22

Cf. Colwell, "Scribal Habits," p. 38l.

135 j
brackets). The instance of alternative conjunctions are also probably secondary. Three of them are singular to P66 j

(7:28 1. ; 9:28 1. 6; 13:29 6e 1. ). Another two involve a pair of alterations (8:23, 25 . . . 1. xat . . . ), the sec

ond of which is singular to P66. The for re in 4:42 is the result either of a misreading or of the substitution of a more common for a less common word. The 1. xat in

8:23 and 9:12 is to be explained as a secondary conformation to a more characteristically Johannine mode of expression. It may be concluded, therefore, that in the varia tions involving conjunctions, ?66 has a tendency to pick up secondary readings, and for the most part these readings again, as in transpositions, create an easier or smoother text. Variations in Verb Forms. Sixty-one items in our list of variations involve verb forms of various kinds. As with the transpositions and conjunctions, the majority of these, when such can be Judged, show secondary processes to be at work in P66. 1. Tense and A k t i o n s a r t . Twenty-three of the 6 l i
|

j
!

v a r i a t i o n s of verb forms involve t e n s e and A k t i o n s a r t . Although a d e c i s i o n based on Johannine s t y l e , o r o t h e r i n t e r n a l c o n s i d e r a t i o n s , i s not s e l f - e v i d e n t in many i n s t a n c e s , t h e r e a r e some which seem q u i t e c l e a r .

136

There are six readings where has a perfect for the aoriet found in the Neutral tradition. Three times the

change is from to (6:57; 7:29; 8:42). Of this verb, Abbott has noted: " is mostly

(15 times) in the aorist, when applied to God as sending Christ.n23 The three readings of are a part of these

15* and are therefore probably secondary. The same is true of the reading of for at 8:42. Abbott notes of or ...

that they "are used for the most part in the aorist

to describe the Son as coming (or being sent) from the Father, but in the perfect to describe His having arrived in the world" (p. 334). Since the context here is ex , the singular reading of is most likely secondary. In fact, 8:42 is the reference Abbott I

chose to illustrate this usage.

There are four places where changes an imperfect to an aorist (9:10, 12; 11:13, 37). Although Judgment is difficult here and must finally rest on whether Johannine Aktlonsarten are meaningful, it is perhaps significant that P66 only once (7:30) varies from the Neutral tradition in reading an imperfect for the aorist, and here it is most
i i
I

3pxammar, p. 332. I

137
Oh

'

likely the result of a scribal slip.** There are two interchanges of the future and present (8:36; 15:21). Again, Judgment is difficult, except to

note that these are singular to P66. Perhaps it is signifi cant that both changes occur in a loglon of Jesus, and in each case the reading of P66 is more "existential" for the scribe's contemporaries. you are indeed free." "If the Son therefore frees you,

"But all these things they are doing

to you because of ray name." There are nine other readings which involve Aktlonsart in moods outside the indicative; and in every instance P66 reads an aorist for the present of the Neutral tradi tion. There exists no full study of Johannine style at

this point. However, there are indications that John uses Aktionsart meaningfully.25 A clear example is the tva The same is true of the (Cf. also,

yvcoTe xat of 10:38.

imperatives in 2:, . . . , notetre. the imperatives in 5:8).

Turner has noted that eav with the present subjunc tive is common in Koine Greek and usually indicates mean ingful Aktionsart, i.e. it "denotes a hypothesis which can
2

^ene0aXXev 1. . The scribe made an iden tical slip at 7:44, but corrected it by rubbing out one . 5cf. Abbott, Grammar, p. 3^9: " . . . John, more than many Greek authors, utilizes the shades of difference between the aorist and present subjunctive."
2

138 ;
occur over and over again" (p. 114). Of with the aorist, he says: "This represents a definite event as

occurring only once in the future, and conceived as taking place before the time of the action of the nain verb" (ibid.) for the most part John appears to keep this meaningfulness in eav-clauses. However, such does not always seem to be the case. Abbott has correctly observed that in John's repeated use of eav with the present, which is more frequent than in the Synoptics, "it is not always easy to perceive the difference of meaning" (p. 371).

P66 four times reads an aorist subjunctive for the present of the Neutral tradition in cav(orav)-clauses (6:62; 7:27; 14:13; 15:4). In two of these the aorist is

clearly the more logical Aktionsart. In 6:62 their "seeing the Son of Man ascending where he was formerly" is scarcely "meaningful" as a repeated occurrence. The same is true of the "coming of the Christ" in 7:27. The more "logical" aorist of P66 in both of these instances is almost certainly secondary. On the other hand, the "meaningfulness" of

Aktionsart in 14:13 and 15:4 would appear to be able to go either way. Two others of the variations of Aktionsart involve the imperative (10:38; 16:24). If our early witnesses are

to be trusted, John has a particular avoidance of the

139 aorist imperative with nioreuetv.2" The morenocr* of F66 So also Is the

In 10:38 therefore is probably secondary. atxi\oaa9t

of P66 in 16:24, which is a much easier reading

than atrctrc. According to Abbott, whose Judgment seezs wellfounded, John in tva-clauses seems to make a "deliberate discrimination in his references to the beginning and the permanent development of 'believing'" (p. 38l). Whether or not this may be styled a Johannine characteristic, it is certainly a characteristic of the Neutral tradition, which reads tva TTtareuiyr with "meaningfulness" (i.e. as the permanent development of believing") at five places (6:29; 13:19; 17:21; 19:35; 20:31). P66 is lacking at 6:29, but

Is faithful to its basic textual tradition in the latter three. At 13:19, however, it reads the aorist, which Is

secondary to Its own tradition, and probably also to John.2? In 13:37 P66 reads the aorist infinitive t

26

Cf. the discussion in Abbott, Grasnar, pp. 3^9-320.

and the Byzantine tradition read the aorist at all times. The present subjunctive at these points seexs to be preferred on the grounds that the aorist Is the acre common reading, and scribes are more likely to have changed toward a more common form than away from It. It seeas less likely that a second century scribe should prefer a less common form and thus create "meaningful" Aktlonsart (the present subjunctive at least has the possibility of being meaningful ) than that he should prefer a aore cooaon expression, thereby disregarding what for the author aay have been "meaningful."

140 for the of and C. There can be little ques tion that deliberate alteration is involved here. In verse 36 Jesus has already said, *t . The question is whether B and C "improve" Johannine Aktlonsart because of the "propinquity of 28 (if so, why here only and not in v. 36?), or whether P66 and most other MSS fail to see what is perhaps a subtle use of Aktlonsart and conform to the first occurrence of the word. If is at fault here, this is assuredly "scholarly1* revision, and to the degree that it is less "scholarly," P66 is here secondary to its basic tradition. The one other variation of Aktlonsart is the reading [sic] 1. at 13:2. Whether the footwashing took place after or during supper can hardly be decided on internal grounds, unless the "dipping of the sop" in v. 26 indicates that dinner was still in progress. Whatever may be the Johannir.e original, P66 is clearly secondary to its basic tradition, being the only early MS of this tradition which reads the aorist participle. 2. Voice. There are five places where P66 differs from the Neutral tradition in the voice of verbs (not including 16:24 above, where tense is also involved).

Por this argument, see H- C. Hoskler, Codex and Its Allies, pp. 351-352.

141 Three of these involve the future of > (6:57, 58; 11:19), which in the NT is far more common in the middle, and is usually universally so attested in the MS tradition. In John, however, the Neutral tradition, often with a few others, almost unanimously supports the future active at 5:25; 6:57; 6:58; 14:19 (only P45 reads the future active at 11:25). At 5:25, where P66 is adhering more closely In the

to its basic tradition, it also reads the active.

other three instances, however, it reads the middle with the majority. The future active of this verb is probably a Johannine feature, since the Neutral MSS do not make a point of changing the middle to the active in the other writings of the NT. P66 therefore is clearly secondary at this point. Although Abbott tries to show distinctions between the active and middle of airetv in John, his distinctions do not seem to hold true. 29 If the middle is used to mean

"ask earnestly," one may well question the undisputed occurrences of the active in 14:13, 14; 15:l6; 16:23, 24. The middle occurs only three times in John, twice with universal H3 support (15:7; 16:26), and once (11:22) where only P45, P66 and W read the active. This latter may be only a scribal slip (the addition of a ) , but it is almost certainly secondary.

^Grammar, PP 389-390.

142 John uses the verb approximately 'Jk times in the aorist, and almost always in the aorist passive, with little textual variation. The aorist middle

occurs at 5:17 and 5:19 in all major MSS except D and W. At 18:34 P66 and several others also read . On

the basis of ardua lectio potior, P66 here may well pre serve the original reading. 3 Mood. There are four readings in P66 where it
i
i

differs from the Neutral tradition in the mood of the verb, and in all four cases P66 has a secondary reading. In 4:14 and 10:5 (where it is Joined by P75) ?66 reads the aorist subjunctive for the future indicative in an emphatic denial. This is probably secondary in the

interest of better Greek. The same is true of the aorist

subjunctive for the future indicative in the t-clause in

3:7.

It is difficult to explain the origin of the indica In 6:6k

tive, if P66 has preserved the original reading.

the alternative in P66 for the of the rest appears to be in the interest of easier Greek. (The future participle is rare

in the New Testament.) Again, to explain the reading of the majority as derived from that of P66 would be difficult indeed. 4. Parataxis. Although parataxis is not as common !

in John as in Mark, it still occurs often enough to make it one of the marks of Johannine style. P66 has two readings

;
i

143
ore evttycv . . . xat . . . xat aveneoev naXtv, ctncv () C W ore evttyev . . . . . . xat avancouuv , etncv Poo A L T p c ore evttycv . . . xat . . . naXtv, etnev D Byz TR tymptov xat to C (L) pc P66 (D A) Byz TR

where the parataxis of the Neutral tradition is broken: 13:12

I [

13:26

P66 is probably secondary in both cases, especially 13:12,. where the participial construction improves an otherwise cumbersome sentence. Possibly the omission of \a\i&avet xat in 13:26 is

also an elimination of parataxis; but it is probably noth ing more than an omission of an unnecessary redundancy.^ (Cf. the singular omission in P66 of . . . xat in 12:11.) At one other point P66 preserves a paratactic construction, which is participial in the Neutral tradition (12:6 , xat 1. ). On the basis of stylistic con

siderations P66 may well preserve the original at this point. In variations involving verb forms, therefore, much

30c. . Barrett, The Gospel According to St John (London, 1962), p. 373* suggests that the \a\i&avet xat may have been added to recall the notable action of Jesus at the last supper, repeated in the Eucharist, of taking the bread before distribution." But by the same token, it may have been omitted because the action here, which pre cipitated the betrayer's foul deed, seemed too closely related to the eucharlstic action.

i
!

144

the same kind of process appears to be at work in P66 as in regard to other grammatical points: P66 sometimes pre serves what appear to be original readings, but at most places where it deviates from its basic textual tradition, it does so in favor of an easier or smoother text. Personal Pronouns. In our list of 376 variants there are 21 which involve the addition/omission of per sonal pronouns. It has already been noted that John has a rather "un-Greek" proclivity toward expressing the personal pronouns in the nominative. There are five instances where

P66 omits the nominative pronoun (6:40; 7:36; 8:42; 13:33; 14:4), and a single instance where it is added (13:34). P66 here seems to show a tendency away from the Johannine toward a more common Greek idiom. On the other hand, there is a tendency in P66 to add the possessive pronoun (e.g. 4:53; 13:22; 18:11; 20:17, 30). Each of these involves or . An omission in such instances would be less easily explained than an addition in the interest of a fuller text. ; The other instances of addition/omission of the per

sonal pronoun involve the direct or indirect object. P66 reads the pronoun at five places (1:38; 9:35; 12:47; 13:36; 14:7), and is without it in six others (7:34, 36; 10:25; 11:32; 13:26; 15:7). 3 1 The readings of P66 at 12:47 and

31This may also be true at 15:22 where the text

1*5 13:26 seem clearly secondary. In 12:47 the pronoun is added to "fill out" the sentence; in 13:26 the omission removes a "Semitic11 redundancy.32 But there seems to be no

clear tendency in P66 at this point, and each instance must be evaluated on its own merits. Moreover, the decision will often be made on external grounds, since the Fourth Gospel itself appears to show no special tendencies at this point.33 The Article Before Proper Names. There are 19 variations in P66 which involve the addition/omission of the article before proper names. In ten instances P66 reads the article against the Neutral KSS (1:35; 6:7* 43; 7:16; 8:12, 58; 9:35; 11:18; 13:29 bis); there are nine instances where P66 is anarthrous (6:70; 9:28, 37; 11:14, 25, 40; 12:2; 14:9; 18:38), not counting nine other places

appears to read without aurmv. But there is an interlinear lacuna directly following . At the lower part of the lacuna there seem to be clear traces of an interlinear addition of the . This reading is therefore included in the discussion of corrections.

32see Barrett, The Gospel, p. 373.

3 3 A check of the remaining nine points of variation against the critical texts of Tischendorf, Westcott-Hort, von Soden, and Nestle-Aland is interesting. Westcott-Hort read with P75 against P66 at all nine.points. Tischendorf and Nestle-Aland read with P75 in every case but 7:3^ and 36, where P75 are almost alone. Von Soden reads with P66 Byz against P75 3 at 7:34, 36; 9:35; 13:36; 15:7. In the other four instances P06 is supported by a very few other HSS. It would seem that the editor's pro clivities toward certain MSS and KS traditions are in evidence here.

146
where the article was originally omitted, but corrected by later addition. To determine the tendencies of P66 at this point is a little more difficult, for here in particular one is not always certain of Johannine style. Eowever, it has often been suggested that P66 tends to support the Neutral tradition with a very high level of agreesent.34 It has often been noted that proper names, espe cially , are perhaps more often anarthrous in John than elsewhere in the NT; but until recently there has been no detailed study of this usage in the Fourth Gospel. Blass-Debrunner-Funk,35 von Soden,3 and Abbott37 have short sections but they only give general directions as to usage, nothing definitive. J. H. Bernard observed that is frequently anarthrous in certain idiosatie

-^This was first noted by Martin in the editlo prlnceps, p. 144. Cf. Smothers, "Papyrus Bodmsr II,;* PP. 430-439, and Birdsall, The 5od~er Papyrus, pp. 15-16. Cf. also Richard C. Nevius, "The use of the Definite Arti cle with Jesus in the Fourth Gospel," NTS, XII (1965/66), . 82-84. The work by Nevius is the most important attempt thus far to deal with this question. Unfortunately, how ever, he fails to consider the evidence of P73- Ee has also Incorrectly cited the evidence for P66 agreement. He has suggested that P66 agrees with In omitting the article at five places where had no previous support (p. 84). However, the inclusion of 10:34 in this list Is quite misleading, since P66 has a complete change of word order and is supported by P45 and W. P75 also supports P66 and at 9:41. ^Greek Grammar, par. 280. 36pie Schrlften, 1:2:319-1. 37Grammar, pp. 57-58.

147
constructions, but his heeitance at this point is revealed in his conclusion: "Where the article is missing before

. the text always calls for scrutiny."38 Much of the difficulty here has been due to the text of B, which is anarthrous in John far more often than are the other great uncialsso much so in fact that the anar throus in John has often been considered to be the anarthrous of Codex Vaticanus. 39 Recently R. C. Nevius has offered a more detailed study of this usage in the Fourth Gospel, in which he noted that P66 often tends to support B.^ He further suggested that the frequent occurrence of an anarthrous text in other early KSS is perhaps a significant clue to Johannine style. His final conclusion seems quite important: "The larger

number of these instances [of anarthrous readings] seems to preclude any rational explanation, other than that we are here confronted with a subtle style which I think must be

3"A Critical and Exegetlcal Commentary on the Gospel According to St;. John (Edinburgh, 1926), I, .43. 39rbjd. Cf. Abbott, Grammar, p. 57, n. 2, who notes that his "statistics" are doubtful owing to . . . the weak nesses of on this point." H. C. Hoskier, whose antago nism toward Hort makes him incautious, charges: "This perpetual slurring of the article before , sometimes by K, sometimes by . . ., is not conducive to a high regard for the care and respect we should expect in these two great manuscripts of antiquity, before whom the schol ars of the world to-day bow down and worship" (Codex B, II, 259, n. 2 ) . ^See supra, p. Ih6, n. 3^

. '." 148 "j traced ultimately to the author of the Fourth Gospel rather than to the scribe of Codex Vatlcanus" (p. 85). Nevius, however, bases this conclusion chiefly on
I

8tati8tic8 of a rather general kind. Apart from his sug gestion that some anarthrous readings may be the result of the beginning of lections, only the anexptOn idiom was isolated in terms of suggesting clues to Johannine style. In order, therefore, to evaluate the text of P66 at this point, an attempt to discover Johannine style is here offered. Although the following study is limited to in the nominative, this is by far the most common occur rence of a proper name in John (approximately 195 instan ces). The results of this study, therefore, should offer

guidelines for a total study of the article before names. The most frequent occurrence of in John is in constructions which introduce direct discourse (approxi mately 99 times). The Johannine idiom at this point has

two basic forms, with some variable patterns within each: etnev and () (-, -) [] Although the second formula is more frequent, there appears to be no set pattern as to when one is preferred to the other. Apart from an occa sional longer formulation involving introductory sentences with participial constructions (e.g. 1:42; 7 2 8 ) , these two basic forms account for every occurrence but three where

proper names are found in introductory formulations in John.1*1

ho

149

In the twenty-fifth edition of Nestle-Aland, * there are 14 occurrences of the full idiom etnev . 3 j n 12:30 (where the is lacking) the

form is broken in P75 L 157 1321 to read xat ctnev . In the 13 remaining occurrences, has an

anarthrous in all but 6:29, where P66 has a lacuna and P75 is anarthrous with W pier. P66 is extant at 13

instances (including 12:30) and is anarthrous in all but 6:43.^ P75 is lacking only at 13:7 and 20:28 and is

anarthrous in every instance where it is extant. There can be little question, therefore, that the basic Neutral tradition has an anarthrous with this idiom. That the idiom is also Johannine, not simply
in

8:12 TTaXtv ; ,12:23 6 ; and 12:44 xat . Cf. 5:17 where P75 W 1241 omit .
ZIP

The use of Nestle-Aland here does not mean to imply that this is the Johannine original. It is simply a useful tool from which to start the discussion. ^31:48, 50; 2:19: 3:3, 10; 4:10, 13; 6:29, 43; 7:21; 8:14; 12:30 (om. ); 13:7; 14:23. There are three other occurrences with another name than (3:9, 2 7 ; 20:28) and seven others with a common noun, pronoun, or no subject expressed (4:17; 7:52; 8:39* 48; 9:30, 34, 36; 18:30). ^P66 is also anarthrous at 10:34 where it reverts back to xat from the "broken" formulation .

150 Neutral, is perhaps demonstrated by the lack of consistency with which the other early uncials read the article. W reads the article only at 7:21 (and 6:43, where it is pre ceded by ). reads the article at four other places:

1:48; 2:19; 3:3, 10 (it breaks the form at 8:14); and D at two others: 4:10; 8:14 (as well as at 6:29 with B, and 6:43 and 7:21 with W ) . All of this seems to Indicate that

this is the idiom of John, and that later scribes, finding it peculiar, tended rather indiscriminately to conform to the more common usage of the article with .
J

This is further confirmed by an examination of the idiom in the instances where it is "broken" into two other basic patterns: and () Co] .46

the conclusion of Nevius, "Definite Article," p. 85: "Indeed, if anarthrous style were a personal idiosyncracy of the scribes of D and B, one might expect to find more consistency in their omissions. There may be some personal preference reflected here, but a case could equally be made for other manuscripts adding the article in a belief that proper names naturally should have the particle." "That is the basic form and the others "broken" patterns seems to be supported by the fact that the MS tradition tends to go toward the former from the latter, but seldom vice versa. P66, e.g., reverts to the apparently basic form at 10:34 and 18:37/ but never goes the other direction. Various uncials add " ] at 3:5; 8:19, 49; 9:13; 13:38; 19:11. On the other hand, the omission of these words in instances where almost all MSS read them occurs only at 3:3 and 12:30 in singular readings of s, and in 13:7 in a singular read ing of Codex 33." This direction of change is quite opposite the

151 The use of the article with is almost identical with that of the basic form. There are 14 occurrences of this pattern (including at 13:8 [ A C L] and at 13:38).^ is

anarthrous in all but two instances (3:5; 18:37), including '19:11 where it adds after . P66 is extant at eleven places (lacking 18:8, 37; 19:11) and is anarthrous in each instance, including 13:8 and 13:36 where it reads and 18:34 where it reads for . 75 has text only at six places (3:5; 8:19, ^9, 5^; 9:3; 11:9) and is always anarthrous. P75, therefore, is consistently anarthrous whenever immediately follows the verb. P66 reads the article once (6:43), plus one other instance where it omits the of the other Neutral MSS (10:25). reads the

article on three occasions (3:5; 6:29; 18:37). The probability that this is a Johannine, not simply a Neutral, phenomenon is again demonstrated by the incon sistency of the other early MSS. W, for example, which

"Atticizing" tendency for which G. D. Kilpatrick has argued in regard to this idiom (see "Atticism and the Text of the Greek New Testament," Neutestamentliche Aufsatze, ed. J. Binzler, et al. [Regensburg, 1963], 126). As far as the Gospel of John is concerned, Kilpatrick has apparently missed the tendency of variation. Cf. the discussion infra, pp. 236-239. 3:5; 8:19, 49, 54; 9:3; 11:9; 13:8, 36, 38; 18:8, 34, 36, 37; 19:11.
47

152 reads the article at 8:19; 13:38; and 18:34, Is anarthrous at both places where is not. Codex Sinaiticus, which reads with W against at 8:19 (hut not at 18:37), also adds the article at 8:49 and 8:54. also has the most

difficulty with the shortened formula, adding xat etnev at 8:19 and 8:49 and the pronoun after at 13:36; 18:34; and 19 J11 where in each instance it also adds the article. Ve may conclude, then, with a high degree of proba bility that the anarthrous when it Immediately follows is a Johannine idiom, and that P66 at this point is faithful to its basic textual tradition; as well as to the original text of John. When this basic formula is "broken," however, by the insertion of either a pronoun or a conjunction (or both) between and , Johannine style is a little more difficult to ascertain. In Nestle-Aland there are twelve such readings.^" B , with P75> omits in 5:19. Otherwise it reads the article only at 6:26; 6:70; In each of these three instances the whole MS

and 10:32.

tradition reads , except s at 6:26 and P66 at 6:70. Moreover, in its eight anarthrous readings is singular at 10:25 and 13:26, is joined by alone at 7:l6, by P66 alone

^5:19; 6:26, 70; 7:l6; 8:34; 10:25, 32, 34; 13:26; 16:31; 18:20, 23.

153
at 8:34, and by P45 V alone at 10:34. There Is therefore

in a clear tendency to be anarthrous even when the Idiom Is "broken." P66 again generally supports B. It has lacunae at

6:26 and 18:23, omits at 10:25 and our at 13:36, aad reverts to the basic idiom at 10:34. Of the reaaining

seven instances it reads the article at 5:19 (with all JE5 except P75 and which omit ), 7:l6 (against ) , and 10:32 (with all known KSS). It has a singular anar

throus at 6:70 and Joins at 8:34; 16:31; and 18:20. Moreover, in two other instances where it reads this for mula (13:8; 13:36) it Is also anarthrous. P75* on the other hand, reads at only four Instances (6:70; 10:25, 32, 34) and has the article in each Instance. (It reads at 8:34.) However, the value of Its witness here is quite limited, since in two of these 3 also reads the article (6:70; 10:32) and in one of the other two Is singular (10:25). The graphic presentation of this discussion is fccodi In Table V H . It will be noted that P66 and are alone In

their tendency to be anarthrous with this "broken" patternHowever, the anarthrous does occur occasionally in the other MSS. The problem is whether one is here dealing with a strictly Johannine idiosyncracy or that of the scribes of P66 and B. Abbott has suggested that "perhaps where or Is inserted, referring back to the

15*
Ot CO V) \

CM

-^

<n

HI

r-4

-? 3

Of

CM

w
- g CM

1 CM

3
CM

s
c-i

Os.

CM CM

CM

9 SB us 5

eo

CO

CM

CM CO Hi

+
n
HI

CM CM

- A

CM
ON

+
I

CO CM CM CO CM CM

a
CM

o
* -O I

CM

CM V\

3
s

c-

HI

V>

CM

CM

s
3 3-~ ** 3 C="CJ W > r v C . W t C J t CJ
IP

rf

nf
UP

*\
IP

s
3 t> c J fc

3 e> cr c > . > =

- 3 * - ' ' W > 3 CJ UP C 3 cr o UP > "** 3 C Q.M t> r >-- t o w* < CJ-

cr > M fcr- "JJ - *>* > w *""* UP * 3 3 w w >- V < 3 c

cr W * - fe >

person spoken to, a"corresponding 6 is more often inserted to refer back to Jesus."^9 However, we shall suspend Judg ment for a moment on this point because a similar problem arises when one examines the other basic idiom, , which is identical to the "broken" formula except for the verb. This basic form occurs 36 times in John.5 There are seven other instances where the pronoun is preceded by ,5* f0U1 others where or occurs without the pronoun," and four others where the pattern has been "broken" to .53 There are only four occasions where appears without a conjunction or pronoun intervening.54 On eight other occasions the simple form without appears, and in all but two (1:51; 21:16), various MSS add (the TR so reads at 4:l6; 6:21; 18:5). An examination of Table VII, as well as an investi gation of particular instances, indicates that the basic

^Grammar, p. 57, n. 2.

5l:43; 2:4, 7; 4:7, 17, 21, 26. 34, 50; 5:8; 6:35; 8:25, 39, ^ 2 , 58; 9:37, 41; 10:6; 11:14, 23, 25, 40, 44; 13:10, 29; 14:6, 9; 20:15, 16, 17, 29; 21:10, 12, 15, 17,22. 516:32, 53; 7:6; 12:35; 13:27; 20:21; 21:5. 527.33. 8:28; 10:7; 12:7. 534:48; 6:67; 8:31; 18:11. 546:10; 9:38; 11:39; 13:31. i

156 Johannlne form is to read the article before . The article before occurs without MS variation at 2:4, 7: 4:7, 34, 50; 5:8; 6:32, 35; 12:35; 21:15, 22. On twelve other occasions an anarthrous is singularly attested (1:43 K*; 4:21 ; 4:26 A; 7:6 K*; 8:39 ; 10:7 ; 11:23 A; 11:25 P66; 12:7 P66*; 13:10 B; 21:10 B; 21:12 B ) . It was with respect to this idiom in particular that was considered to be at fault. Abbott suggested that the peculiarities of here were probably the result of a con fusion of the final of with the article,55 which in the uncials are so similar. This, however, cannot be

the explanation for the same phenomenon in P66 and P75 P75* for example, which reads the idiom at 30 places, varies from only at 6:53; 7:35 (P75 singular); 8:39 (B singular); 8:42; and 10:7 (B singular). At 6:53 and 8:42

is Joined only by P66. The other four anarthrous read ings in P75 (8:25, 58; 9:4l; 11:44) are all supported by B. This means, therefore, that although they do not entirely agree, P75 and both witness to a similar phenomenon, which is not a peculiarity of Vaticanus alone. The witness of P66 further confirms this. P66

shares 37 readings with B, of which six are anarthrous (6:53; 8:25, 42; 9:4l; 20:15, 29). It reads an anarthrous against at seven other points (9:37; 11:14, 25,

Grammar, p. 57, n. 1.

157 40; 12:7; 14:6, 9 ) , and has the article where is anar throus at seven others (8:39, 58; 10:7; 11:44; 13:10, 27, 29). Although this does not amount to full agreement, it

does point to a similar phenomenon occurring in both MSS. It is also interesting that this phenomenon does not occur in the Neutral tradition until 6:53, and thereafter occurs in earnest from chapter 8 on. Of the Johannine use of the article before in constructions introducing direct discourse, we may make the following general conclusions: 1 In the idiom xat ,

it is the Johannine habit to have an anarthrous when the noun immediately follows the verb. This does not mean, of course, that it is an ironclad rule. 2. In the idiom, where the verb and

noun are separated by the insertion of a pronoun, and in the OUTUJ Co] idiom, the Johannine habit is to read the article before , with occasional lapses to an anarthrous reading. It does not appear that any pattern

may be established as to when the reading is anarthrous. The other 93 occurrences (as counted in NestleAland) of in the Fourth Gospel are not as easily classified. However, the above analysis indicates that a clue may be found in terms of word order.5 The 93

Of. E. C. Colwell, "A Definite Rule for the Use of

.158 occurrences were accordingly classified into three basic word-order groupings: (l) where precedes the verb

(26 times), (2) where immediately follows the verb (42 times), (3) where follows the verb, but is separated from it by other words (25 times). An analysis of the readings in these groups yields the following results. 1. In most of the places where precedes the

verb,5 the Johannine usage is clearly discernible. On seven occasions it appears at the beginning of a sentence with the common Greek idiom, or On every other occasion is anarthrous in the Neutral tradition and usually is so in the entire Greek tradition. It is always anarthrous when It Is the first word in a sentence and is followed by the conjunction (6:15; 8:59* 11:33* 38; 12:44; 18:4; 19:26). On nine other occasions,

It Is the first word in a ort -clause. The KS tradition here, although not unanimous, is overwhelmingly in favor of

the Article in the Greek New Testament," JBL, IZI (1933), p. 13, where he suggests that his "rule" was discovered on the basis of word order. 57ThIs means participles as well as main verbs. It was found that when is used with a participle and a verb, and the participle precedes the verb, usage of the article with appeared to be controlled by Its relationship with the participle rather than the verb.

159
the anarthrous usage as Johannine.5 on three other

occasions is accompanied by the intensive (2:24; 4:2, 44). Except for 2:24, the MS tradition is strongly in favor of an anarthrous here as a Johannine idiom. It is interesting to note further that when

ever any of the major MSS invert the word order from verbsubject to subject-verb, they invariably keep the Johannine idiom. Thus P66 D read for

at 11:30; D for at 11:51 A L 33 PC for at 12:12; and ?66 for at 19:28. This means, therefore, that except for the common bt idiom, is anarthrous in John when it precedes the verb. P66 at this point conforms both to the Johannine idiom and to its basic tradition, except at 4:47, where the scribe rubbed out an original o, apparently correcting his own error. 2. In the 25 instances where follows the

verb, but is separated from it by another word, the article ie read without variation in the majority of instances.

58ln two of these (20:14; 21:4) is the predicate noun. The usage here, therefore, conforms not only to this Johannine habit, but also to "Colwell's rule" that proper names are anarthrous when they are predicate nouns. See "Definite Rule," p. 20. This also explains the only instance (6:42) where is anarthrous without MS variation, when it follows the verb.

160 is singularly anarthrous at 6:5 and 6:61, L at 6:17, D at 11:4, and at 5:14 and 19:5. Besides these, is anarthrous In the following four Instances: 6:3 12:1 P66 * D W A L rell 6e ro 6e

66 A D X 19:30 V * 21:1

13 124 eyetpev ex cyetpev ex V Lpn ov ex Byz TR ex L Byz TR

C A 0 Byz TR D () pc

This means that, as elsewhere, P66 tends to support In its anarthrous readings (failing to do so only at 5:14). P75 has lacunae at each of these instances, except 5:14 where it supports P66. The Johannine habit seems quite clear here: Where is separated from the verb, the author almost always reads the article. However, the strong Neutral and Western evidence for an anarthrous at 6:3 nay indicate that the author himself occa sionally deviated from his normal pattern. 3. More difficult of solution are the instances where isnediately follows the verb. Whereas there is only one instance where all HSS are anarthrous (6:42, where it is a predicate noun), there are ten where all agree in reading the article (2:22; 4:54; 5:6; 6:1; 7:37; 9:14; 11:30; 19:28; 20:2; 21:25) and seven others where

l6l only one MS Is anarthrous (7:1 B; 10:23 B; 19:20 X; 20:26 W; 20:30 D; 21:7 .D; 21:20 D ) . It will be noted from Table VII that has more anarthrous readings here than any other MS. However, many of these occur In the later chapters of John, where P75 is lacking and P66 has many lacunae. P75 and share 20 read ings where Immediately follows the verb. Of these they share six anarthrous readings (5:1; 7:14; 9:35; 11:32, 51; 12:16). F75 and disagree three times where the

latter is anarthrous (1:47; 7:1 sol; 12:36), and once where the former is anarthrous (7:28). Much the same situation prevails between P66 and B. They share nice anarthrous readings (5:1; 7:14; 11:32, 46, 51; 12:16, 36; 13:21, 23); they disagree once where P66 is anarthrous (11:35) and four times where is anarthrous (1:47; 7:1 sol; 9:35; 10:23 sol). Both this high level of agreement at a rather insignificant point and the frequent instances where P66 P75 are joined by other than Neutral MSS seems to indicate that, although with the article is the more typically Johannine pattern, an occa sional anarthrous when it immediately follows the verb belongs to the author and not to the scribes of the Neutral tradition. It should be noted by way of conclusion to this discussion that, although P66 does add/omit the article sometimes at variance with the Neutral MSS, for the most

162 part it adheres quite closelyalmost rigidlyto its basic textual tradition. Moreover, in so doing it is generally

in keeping with Johannine style. In fact the witness of P66 may be very important for the discovery of Johannine style at this point, since this is one of the few points of grammar where P66 fails to pick up a pattern of secondary readings, in the sense that it fails to read the much more common Greek idiom of using the article before the proper name. Miscellaneous Variations. There are a few other points of Johannine grammar where variations in P66 may be checked, and where its idiosyncracies may be noted. At three places P66 reads the attracted form of the relative pronoun against the Neutral tradition (4:5, 50; 7:39). It fails to do so at 2:22. That the non-attracted

form is not merely a Neutral phenomenon is certain in that the Neutral MSS do not frequently so read outside of John. P66, therefore, is probably again picking up secondary readings, both in terms of its basic tradition and of the Johannine original. Another Johannine characteristic is the use of ex with the genitive for the partitive genitive.59 At this

point P66 has a good record with respect to the Neutral tradition, eliminating the ex only at 12:9.

59see Schweizer, Ego Eimi, p. 92.

163 P66 shows several non-Neutral propensities where variant spellings are involved. It seems to have a tend

ency to read ev for (3:27 with P75 B; 5:19, 30; but cf. 1:3). It favors for (l:32; 6:10j 19:39), for ovrtw (7:8, 30), and prefers the declinable form of papta to (11:19, 28, 31, 32, 45; 20:11; it has been corrected, 11:32 and 20:11). Among its non-corrected singular readings of all kinds, there are many (besides those noted in the above sections) which seem to involve the secondary processes of smoothing out, harmonizing, or of being "more instructive." The following are but a sampling: 1:50 v. 46. 6:6l The redundant is probably a careless P66 reads to harmonize with

harmonization to the ordinary Johannine etnev . 6:69 The singular (with sa)

is probably a partial harmonization to Matthew, which later scribes took the whole way.
;

10:16

is more expressive than the

of the rest. 11:20 The addition of seems to be a curious

emphasis on the fact that Mary remained at home, while Martha and Jesus were outside the village. 11:27 The addition of appears to be in the

interest of a more direct answer to Jesus question.

164
13:5 is a more expressiveand accu rateword at this point than Ytrirnpa. 13:33-4 The addition of the non-Johannine and

the mark of punctuation before it, is clearly in the interest of making better sense of the text. 15:13 The for is probably for emphasis. There are a number of other variants in our list, which often involve considerable differences of meaning in the text. Out of some 26 that have been so assessed, only one (16:27 . ) has found its way into the Nestle-Aland text. While this is not a sure criterion as to whether a reading is original or secondary, an examina tion of these variants on the basis of "critique rationnelle," seems to support the Nestle-Aland text against P66. A few instances follow, in all of which P66 is the (The reading

earliest Greek witness to the text of the TR. of P66 TR is always given first.) 3:25 . .

Of this variation

Barrett correctly observes:

" . . . the singular is unique

in John, and is more likely to have been changed into the plural than vice versa.""0 In fact there is scarcely any

explanation for the singular, except as a deliberate change reflecting the Sltz lm Leben of a second century scribe.

"The Gospel According to St John, p. 184.

6:42

1. . This of P66 and the TR is sec

ondary on all counts. Whether it is an inadvertant scribal error or a deliberate change, the direction of the change is surely to read the more common for at a point like this. 8:38 i 1 . . The reading of =

is more likely due to an assimilation to the preceding clause, than is the well-attested a deliberate attempt at variation. One must ultimately ask at a point like this, to whom is one to attribute the greater Insight, to the author or to a subsequent scribe? Distinctions are

probably to be made between the two verbs, and such distinc tions probably belong to the author, not to a second century reviser. 11:31 . . Again, the uncommon could hardly be explained were the frequently appearing original. 11:57 . . Although either reading i

is admissable, it seems most likely that a change would be made from the generalized "orders" or "directions" in favor of a specific "order." 13:18 1. . The of P66 TR presents a ;

smoother text and is much easier to account for as an alteration of , 14:16 than vice versa.

. . Again , which is probably an

assimilation to v. 17, is more easily explained as a change

166
t

from than vice versa. The results of this study, therefore, Indicate that, although P66 is basically a member of the Neutral tradition, it has a strong strain of readings away from this tradition, which for the most part are secondary both to the tradition and to the original text of John. A large part of these secondary readings are in favor of a smoother, easier text, the kind of corruption that is predominant in the Byzantine MSS of later centuries. These conclusions are almost iden tical to those made earlier by Birdsall, whose conclusions seem worth repeating: In these examples of an attempt to assess P66 by gram matical and stylistic criteria we but emphasizein an acceptable sense--that this is a very 'mixed' text. It is a mixture of good and bad, of primitive and recensional. We find in the Codex acceptable readings . . . side by side with patently secondary readings. . . . Very few of its singular or subsingular readings com mend themselves as possessing a prima facie claim to originality. Collation of its readings with the extant evidence and examination of them in the light of intrinsic criteria of style and language emphasize insights already apparent from the slightly younger P45. In these third-century manuscripts, whose evidence takes us back into the mid-second century at least, we find no pristine purity, no unsullied ancestors of Vaticanus but marred and fallen representatives of the original text. Features of all the main texts isolated by Hort or von Soden are here foundvery differently 'mingled1 in P66 and P45.l

The Bodmer Papyrus, p. 17-

167
II THE CORRECTIONS OF P66 One of the most important features of P66 la the large number of corrections to its text (approximately go 450) Although the majority of these are corrections of

scribal slips (approximately 235), a large number.of them involve alterations to the text in which both the original
i

and the corrected readings are shared by other important MSS. It is these alterations which are our primary concern in this chapter. However, before examining them, soaae preliminary considerations about the corrections are in order. Pie Nature of the Corrections. The corrections are of four kinds: (l) addition, (2) deletions, (3) correc

tions of word order, a;nd (4) alterations involving deletion and rewriting. Both their quantity and the nature of the errors lead to one conclusion: The scribe was a careless and ineffective workman. 3 g e falls into almost all of the common scribal errors, such as dittography (1:27 G V T O U 62 Finality of Judgment is difficult here. There are some places where the scribe has clearly written a letter incorrectly and crossed it out (e.g. 1:19 []). At other places one cannot tell whether a blank space is the result of the crossing out of a letter, or simply a rough spot on the papyrus (e.g. 2:12 ). " c f . the Judgment of V.%Martin, Papyrus Bodser H , p. 30: "Leur abondance incite a les attribuer a l'inattention du scribe . . .," and E. C. Colwell, "Scribal Eabits," p. 386: "Wildness in copying is the outstanding character istic of P66."

168 ; 12:26 eav euot ), haplography due to homoioteleuton (4:49 . . ; 10:10 om. . ), or far more commonly, the dropping of a letter op syllable. Other corrections Indicate that he was fre quently copying without paying attention to the context.** Moat of the simple scribal errors have been correc ted, ^ an(j there seems no good reason to question the Judgeent of Martin that "la reparation de ces tree nombreuses omissions Italt selon toute probability due au scribe original. En tout cas rien n*oblige & les attribuer & une autre main.""" At many points this is clearly to be seen: e.g., at 3*3 where the has been scraped and followed by ; at 4:11 where the of an original was crossed out and followed by the correct ; and at 14:12, where the scribe apparently started a dittography of the preceding , caught himself after he had written , then

**See e.g. the when Jesus is speaking to Nicodemue alone; the mark of abbreviation over the of (4:11), which means he started to abbreviate one of the nomlna sacra; and the TUB for op (4:217: ->I note the following uncorrected readings: 1:4 oa. ev; 2:11 [] ; 4:23 ; 4:1 ; 4:39 []; 5'30 ; 5:36 []; 6:52 ; 6:53 ; 7*24 [ ] ; :20 [ ] ; 9:17 []; 11:35 ; 11:38 ; 13:2 ; 13:9 CYW 3; 18:30 . &&V. Martin, Papyrus Bodmer II, p. 31.

3L69 j crossed out the and changed the t to o. The many dele

tions where a scraped letter has been left blank are also of this type. The Important thing to note here Is the frequency of omissions, which range from one letter to several words. Since so many of the "scribal errors" are of this type, and since so many of the corrections where P66* has a singular reading are also of this type, one may well hesitate before attributing such singular readings to anything more than the carelessness of the scribe himself. A case in point Is the list of forty-nine readings for which M.-E. Boismard found support in the versions and Fathers (supra, pp. 77-78); for over half of these "singu lar" readings have been corrected, and most of these involve the omission of a letter or syllable. Finding textual relationships in such readings seems to be a doubt ful procedure. But while inattention or careless copying is the probable explanation for most of the corrections, there are others for which the only plausible hypothesis is that the MS was corrected against another MS (or MSS). This is j

especially true of the corrections involving the addition, deletion, or alteration of significant words or groups of words. One of the difficulties here is whether these cor rections were made by the original scribe or by a subsequent i j

170 hand(s). Martin was duly hesitant at this point;7 and the

present writer was perhaps too ambitious in assuming that only the original scribe made corrections to the MS. The

full investigation of this question should be undertaken by an experienced paleographer; however, some observations which do not necessarily require expertise may be in order here. The proper place to start such an investigation is probably with the additions, where the scribe*s hand is in evidence. The letters of many of the additions are not as well-formed as those of the original text, but for the most part they are not so different as to suggest a second hand. Probably any differences are the result of the scribe's being less a calligrapher when he has turned corrector: the letters appear to reflect more haste, and they are usually smaller. However, there is one correction which seems clearly to be the work of a second hand: the addition of !

at 13:19. The square and simply demand a second hand: in the original scribe's hand (even in the

corrections which are obviously his) these letters are

j

67
Papyrus Bodmer II, p. 32.
68

I
j

"Early Textual Transmission," p. 248.

171 .
always well-rounded. ^ Although this particular hand does not seem to be clearly in evidence at any other point, it doe indicate that a second hand has had access to the MS. More difficulty arises with the deletions. They are chiefly of three kinds: scraping out, crossing out, and the special mark of deletion [()} above the word. How ever, the original scribe himself was not consistent, deleting first by one method and then another, and some times by a combination of marks. For example, a nevre at 12:1 is deleted with the mark () over the nev, but the TC has been scraped out and written over it. He scrapes out the at 3:3 and follows it with oot, but crosses out the of the incorrect at 4:11 and follows it with an
OD.

The fact that the scribe is notoriously inconsistent and that each of the separate kinds of deletion may at some point be demonstrated to be the work of the original scribe, make difficult any possibility of detecting a second hand. However, there is one point at which such detection may be possible: the use of the special mark () Sometimes

this takes the form of a single dot over each letter to be deleted (1:27, 29; 2:2; 9:36; 10:33; 11:7; 12:28, 40;

is also one of the only two corrections of this length which is interlinear rather than marginal. The other one (15:10), however, is clearly the work of the original scribe.

13*21) and sometimes it takes the form of a series of dotsalmost dashesover the whole word (or words), with out regard for the individual letters (1:49; 6:58; 7:39* 40 M s , 46; 8:33; 10:7, 9, 26, 39; 11:33; 14:3, 4; 16:25, 32). Perhaps this indicates a second hand at work, but one

cannot be sure. And since the scribe is inconsistent in other ways, he may well have been at this point as well. One of the major corrections in the MS, and one which may help in finding a solution to the difficulties at this point, is 11:33. By a combination of scraping out some words and writing over them, by leaving the no fTvT intact, and by deleting the final of what must origi nally have been with the special mark ('*), the MS has been changed from [3] - [ ], shared by P75 et al., to , shared by P45 D et al. The letters of the words written in over the scraped portions have all the appearances of having been written by the original scribe. They are clearly not the work of the corrector who added the portion at 13:9. What slight deviations in letters do appear are probably nothing more than the result of writing over the papyrus where it has become rough from scraping. Since this major change to the MS was probably done by the original scribe, and since there appear to be very few corrections which are definitely the work of a second hand, one may safely assume that the major part of the

173
corrections was aade by the original scribe. Furthermore, the large number of corrections such as 11:33 above. Involving variants which are widely attested In other early MSS, means that we may conclude quite positively that the scribe of P66, after copying from one MS, had opportunity at a later time to check his copy against another MS, with the result that in a number of Instances he chose one reading over another and changed hie own MS. Textual Relationships of the Corrections. The primary Initial Interest In the corrections, as in the original text of the MS itself, was to determine their textual relationships. From a rather Incomplete list of corrections, A. F. J. Klijn concluded that " . . . in almost all cases the original uncorrected reading Is of a Western, In any case non-neutral, type," and that "the corrections are commonly In agreement with the 'Egyptian ."70 This

was the most coaaon appraisal of the corrections, and has recently been advanced by Metzger in his handbook.'* The

present writer has taken some exception to this point of

"Papyrus Bodmer II," p. 334.

7lThe Text of the New Testament, p. 40. "It is interesting that in some twenty cases where the copyist has made alterations between the lines and in the margins, the deleted text almost invariably belongs to the Western tradition, and the reading which the copyist preferred belongs to the Alexandrian type of text.

17* view by showing that, although in terms of individual MSS P66 Boat often abandons a reading it shares with D and more often corrects to read with P75, there is no clear indica72 tion of one textual tradition influencing another.' However, further examination of the MS for this present study, plus the conclusions of Chapter II about the nature of & in John, has indicated that another analysis of the corrections in terms of textual relationships is needed. Moreover, because so many of the corrections are of scribal errors, the canon "to weigh before counting" seemed particularly appropriate here. For this reason, the following analysis is not concerned with most of the singular readings of P66* nor with most of the readings which it shares with a single late uncial or a few isolated cursives. 73 The analysis is thus limited to the > corrections X in P66, where, for the most part, both the original and corrected readings are supported by other MSS. For the purpose of quickly ascertaining the textual relationships of these corrections, they have been conveniently listed In the six following groupings: 1. Corrections where P66* has a Western reading,

72"Early Textual Transmission," pp. 254 ff. 73rhe list of corrected singulars and those Judged to be sub-singular will be found in Appendix II.

175
and P66c reads with the Neutral tradition, usually in agreement with the Byzantine MSS as well: 2:11 4:1 6:5 P66* (*) f q (b) wait P66c P75 A W rell 66* KD 565 1241 1293 itPl P66c P 7 5 B A C W L B y z f q T R 66* D aur a b e f ff2 1 r1 vg P66c P75 B A W L Byz e q sa bo TR

6:64 omit P66* e syc. P66c rell : 7:12 omit P66 D a c e ff2 1 syC p. (P66c) P75 V L 029 ante nept Byz TR 7:14 66* D 565 P66c P75 L Byz TR W a b q r 1 P66*vi<* i e 22 69 138 543 1216 it vg > P66C P75 V L X Byz TR

7:37

7:37 omit P66* K D b e P66c P75 W L rell 7:40 TODY 6* D P66c P75 B L H T T l p c a e G b f q r* boP^ aur c ff2 1 vg W 122 127 229 syC S 2 A pier TR 157 64 7:46 66* * D P66c P75 B W L N T X 3 3 P C Byz TR 9:18 omit 66* 565 660 itPl sy3 bo * p66c P75 B K A D W L 6 N Byz t TR 10:6 66* 1170 lat P66c P75 B M D V L Byz TR

176
10:36 12:3 P66* P45 vld R D W 0 28 69 124 1093 PC P66c P75 A B L Byz TR P66* P66c P66* P66c (D) b c e ff2 r1 B U V Rell aur a f vg TR D W 71 185 482 1093 b ff2 1 r1 vg P75 A L Byz a aur c e f TR

omit 12:31 omit 12:40 omit 17:14 omit

66* D a e f 1 _ P66c P75 A W L aur b c ff2 q r1 rell TR P66 D b r1 sy8 P66c M 248 253 > A C W rell

18:40 omit P66* b c e ff2 r1 P66c B U V DSUPP Byz a aur f q TR 19:5 omit P66 a e ff2 r1 xat : P66c S W rell [Note: This reading has been incorrectly cited in MartinBarns. They read a mark of punctuation after ; but it eeems clearly to be this scribe's special.mark for addition, ( / ) This is also the opinion of Aland. See . "Neue Neutestamentliche Papyri II," p. 70.] 2. Corrections where P66c reads with the Neutral MSS against most of the rest: 2:15 ro 4:52 xat P66* Byz TR P66c 75 W L 083 33 213 579 bo b q 66* 0125 2145 e bo sy bo0 P66c P75 C W L U 33 579 124l PC D Byz TR

6:55 *0 66* K* D Byz TR ?66c P75 C W L F 029 33 579 PC bo 7:39 66* V L X N Byz TR P66<> P75 42 91 053 254 e q D f _ aur a b c ff2 1 rl vg sy 7:40 66* Byz f q TR omit P66c P75 8 D W L X 1 565 PC itP1

177
7*41 6* K'D T By? TR * oc P66 P75 W L 213 1071 1170 pc 7*46 " 66# * ay8 Byz TR D omit " P66c P75 W L KC 225 10:22 P66 A D X Byz TR aur c f ff2 P66c P75 W L 33 579 892 1071 1321 omit 565 251 1010 a b 1 10:26 () (66*) A D Byz TR omit P66c P75 W L pc 10:28 atuivtov t 66* A D Byz TR aiuuvtov P66c P75 W L pc 11:29 omit 66* A D Byz TO be P66c P75 C* W L 053 213 33 579 PC 14:4 xat 66* A D Byz TR P66c C* W Q L X 33 579 PC 3. Corrections where P66* has a reading shared by the Neutral MSS and where P66c has a reading often found later in the Byzantine tradition. It will be noted that a great many of the readings of P66* in this list are strictly Neutral readings:
;

2:12 2:12 6* P66c 4:25 66* F66c 6:40

1:42

66* P75 L bo P66c Q A W Byz TR 579 66* 75 L 083 0162 c P66c A W Byz itP1 TR 75 V L Byz TR A F Q 2 565 1241 124 pc b bo 75 B K * A C D W 9 T Byz TR Kc L 0 33 1241 213 1071 PC

66* P75 A C D W L Byz TR P66c U

178
7*52 8:21 8:25 8:28 9:36 ex () ex omit omit omit (66) 75 L T p c 66 D W Byz it TR

P66* P75 P39 D W L b e P66c Byz itPl TR P66* 75 476 66<> D W L Byz TR

66* W L 1 565 1241 a P66c 75 D Byz TR 66* 75 W P66c A D L X Byz TR P66* P75 W L 579 892 1241 P66c P45 A D 33 Byz TR P66* P75 A L Q Byz P66 c D W A 33 579 1241 565 pm TR P66* P75 D V L X 0 33 565 PC P66 c 579 b c 1 rl A Byz TR 66* P 7 5 B K A D W L B y z b f l T R P66c U 28 348 117P 1242 1279 it? 1 66* L P66 c A D C W r e l l TR 66 P66 c U C D H r e l l TR P66*vid C L X pc P66 c A (D) W Byz TR

11:54 epeivev 12:16

12:26 eav eav 6c xat eav 12:26 13:21 13:23 13:24 13:25 14:11

xat Xcyet etne ein entneoojv

66* B C* L * 33 892 pc p66 c K* A D W Byz TR

P66* P75 229 sa p66 c A D W Q L Byz TR 579 P66* P75 V 579 a 66 A D Q L X r e l l TR

14:17 14:22

Ytveoxet ytvcKJxct omit xat

P66* P75 A D L 33 700 544 1071 1355 P66 c W Q Byz TR

179
19:*
# at 66c * B A L X K l I 3 3 544pc *ln\9tr ouv P66 W Byz TR Dupp r 1 209 213 291 1346 itP*

20:18

P66* C* 078 pc P66C K D L Byz TR W Q 33 pc

4. Corrections where P66 c has a Western readings P66* P75 A C W P66 c D L 69 124 33 579 a b e q Byz TR 185 1170 c d f ff 2 1 11:33 () eauror (P66* P75) ( A) C L Byz TR P66 c P45 D 1 131 22 660 sa 11:41 66 75 A C W L Byz TR P66 c D 28 33 69 PC ItP 1 12:47 xat 66* 75 A L Byz aur q vg TR xat P66c D S 0124 pc a b c e t ff2 1 r 1 5 Corrections where P66 c reads alone against the 4:51

rest: 3:33 alt P66* P75 B M D V c P66 c * P66 2 alt ctnoY rell

8:25 12:22

P66 P75 S D rell P66 c

xat . 6e xat . 66* . 6e xat . P66 c xat epxcrat . xat . 157 xat naXtY . xat . (w) Byz TR . xat . D epxcTat . xat . 75 A L pc 13:24 66* A D V rell cine P66 c
14:14 cr
ere

P66 D V Q Byz TR 75 c A L 060 33 124 1071 1093 PC P66 1241

160 14:26 omit 66 A D V rell P66 c sy 8 6. Corrections where the early MSS are divided, and the variants do not seem to fit categories such as "Neutral" or "Western": 1:19 omit P66 P75 * L Byz TR p. Aemraq P66c A X 157 579 pc p. anearetXov C 33 892 1Q71 pc 66* U C L f Byz IS P66c P75 * 157 c f rl

1:22

1:27 etpt P66* P75 C p5 L 565 pc aur q etpt C Y * P 6 6 C 68 33 579 PC * tpt Byz itPl vg TR 1:36 3:34 5:29 at 66 C* 892 a pc omit P66c P75 A TR 66* 12 40 63 253 254 945 1223 P66c P 7 5 B K A C D L 0 8 3 B y z T R

x<*t ot P66* ot P66c B a e f f 2 ot 6c P75 * A D L Byz itP* T R OTt eepenr OTt tQtwpmr
*

6:2

66* Byz TR P66c D L 053 33 69 pc P75

6:10

ot ot

66 D L 33 579 1241 565 P66 c ? Byz TR *

6:44

CY P66* A C D L 029 Byz ffi omit P66c P75 892 270 251 l6o4 7:4 66 D* sa bo P66 c P75 L Byz it vg TR 8:54 66* A L Byz TR P66c P75 * D C* 579 713
8:54 66* B * K D F X Y 1 3 346 892 1071 P66c P 7 5 A C L H 8 B y z T R

181 9*15 9:17 omit xat 66 V U 053 69 124 892 1241 pc 66 B K A D V L N e B y z T R

oeauTou 66* 75 53 247 bo P66 c B A V Byz TO caurou D P66 * P66 c B A C D W L B y z T R

11:35

11:45 12:9

6 P6 P45,K A* V L Byz TR P66c 0l4l 11 242 314 473 pc C D A c 1 244 249 1010 1293 P66* P75 Q X B c 068 Byz P66c W 1010 * L 047 579 892 1241 28 pc 0 it sa 66 L 1216 1219 660 ?66 6 2 A D W Byz TR

6c 12:12

P66* V b c e f f2 1 c TOTC P66 P75BKADLQByzTO 12:18 P66* P75 185 245 PC xat P66 c A L Q X Byz TR > xat 12:16 omit n*ep imep 14:2 , omit or 14:17 eortr orat 18:34 18:40 omit 12:43 P66* P75 A D Byz TR P66 c W L X V 33 565 579 pc P66* Byz T R c P66 0 5 5 6 5 p66* D V 1 22 69 251 254 5^5 660 itP* P66 c U Q L 8 T Byz TR P66* K* P60 DSUPP 59 472 a aur c f ff 2 1 P66 c B A C V L K 8 T Byz b e q i B 66* D^PP P66 c B A H Byz TR
66**1* * V P66c P60 (A) Byz TR 3>supp pm K c L X pc

19:15 ot 6 ot 6e ot 6e ouv exetvot

182 It seems olear from these lists that P66c shows no special proclivities toward or away from a given textual tradition. For although the corrections are "commonly in

agreement with the 'Egyptian MSS" (approximately 31 times), they are Just as often away from that tradition (approximately 32 times). The closest thing to one clearly defined tradition influencing another is in chapter J, where in a series of ten straight corrections the MS is brought into conformity with the text of P75 ( w . 12, 14, 37 bis, 39* ^0 bis, 41, 46 bis). Six of these readings were a part of the "mix

ture " of Western readings, noted above to be more heavily concentrated here than elsewhere in the MS. This series of corrections is finally broken at 7:52 by a change of word order, almost certainly from a second VorlaKe,74 where the scribe abandons a strictly Neutral word order for that shared by the Western and Byzantine MSS. Besides this, there are only two other series of corrections which may reflect the influence of a definite textual tradition. There are three significant corrections

in a row in chapter 10 ( w . 22, 26, 28) which correct toward the Neutral MSS, and three in chapter 11 ( w . 33 #

'^This change involves the now famous article before . It appears, however, that there has been a half hearted attempt to delete it. Such a deletion and word order change are almost surely the result of reading a second Vorlage.

183
!

4l* 5*0 which abandon this tradition. Elsewhere the corrections are "mixed" in about the sane proportion as is the original text itself. For example, in chapter 12, four "Western" readings are corrected ( w . 3, 16, 31* ^0), but these are interspersed with three corrections which abandon strictly Neutral readings ( w . 16, 26 bis), as well as two others where the MSS are divided ( w . 18, 43) and one where the scribe abandons a reading shared by the Byzantine MSS in favor of a singular reading (v. 22). Except for the three places noted above, it is this type of "mixture" that prevails throughout the MS. . With the lone exception, therefore, of the series of

corrections in chapter 7, and perhaps those in chapters 10 and 11, the conclusion made formerly by the present writer seems valid: " . . . the statistics . . . seem to indicate

that here (in P66 c ) we are dealing with another 'mixed* text, not with a clearly identified tradition."75 One should perhaps be wary at this point of the rather facile assumption that the corrections were all made from a single second Vorlage. If there were only one other

MS against which the corrections were made, then the conclusion is valid that it was "mixed" in about the same proportion as the original Vorlapse. But the possibility

75"Early Textual Transmission," p. 256.

184 !

always remains of more than one other Vorlage, which makes . conclusions at this point tentative. But whatever one may conclude as to the nature of the second Vorlage(n), of the scribe himself it may be con- eluded that he as often chose a reading away from the Neutral tradition as toward it, although he did more often choose to read away from than toward the Western tradition. The important question which remains is to determine whether one can ascertain the principles on which this "editorial" activity was predicated. Textual Characteristics of the Corrections. In order to evaluate the textual characteristics of the correc tions, they have been classified and examined under the same general grammatical criteria used in examining the

original text of P66. 1. There are four corrections Involving word order (6:5; 7:46, 52; 10:28). Although none of these seems to

involve points of Johannine style, one does follow the tendency in the MS of placing the subject before the verb (7:46, see supra, p. 125). On internal grounds alone, however, none of these may be Judged as original or secondary. 2, In corrections involving conjunctions, the

scribe appears to "editorialize" after the faehion of the original MS. There are four places where he removes asyn deton by adding a conjunction (9:10; 11:29; 12:26; 14:22).

185
Although euch additions are not necessarily secondary in terns of the Joharmlne original, they do reflect a process of smoothing out the text. The same is true of two of the alterations involving different conjunctions, from ot 6 to the more cooraon our in 7:30 and from the paratactic at to our in 19:4. However, there are three changes which do not fit this pattern. The change from 6c to rorc may be for stylistic reasons, but it is not a change toward an "easier" text; and the deletion of the xat (shared with W) at 5:29 and the singular at 5:43 are inexplicable on the principle of creating a smoother text. But it should be noted that these are both singular (or nearly so) in P66 in a section where it maintains a high percentage of relation ship with P75 B. Perhaps, therefore, they are corrections

of "slips" from the original, basically Neutral Vorlage "slips" of a kind of which the scribe has a rather large number. It should be noted also that he corrects a singular (or nearly so) asyndetic text at six places (3:21; 4:48, 52; 10:7; 12:2; 13:5), leaving only one instance of singu lar asyndeton in the entire MS (14:9 om. 6c with a ) . Considering the frequent number of omissions in the papy rus, this single instance of singular asyndeton is perhaps a remarkable record, and seems to substantiate what has been noted above: in the matter of conjunctions the scribe shows a definite tendency to smooth out the text.

3 Vbat la true regarding conjunctions is also true about pronouns. There are ten Instances where P66c adds a pronoun to the text (1:22, 27, 42; 2:12; 8:28; 11:41; 12:26, 31; 14:17; 18:34), and a single Instance of deletion (7:40). But this deletion eliminates a rather awkward "these his," and is clearly in favor of an easier text. There are besides these, thirteen instances where the singular oaission of a pronoun has been corrected (2:20; 3:33; 6:52, 60; 8:46"; 9:39; 11:5; 14:26; 15:10, 16, 22, 25; 18:2). At only one place has the scribe deleted a possessive pronoun which is singular to his text (9:8), and this probably reflects the iaaedlate correction of an error. Vith respect to personal pronouns, therefore, both the original text and the corrections In P66 show a clear pattern of picking up secondary readings. 4. There are seven corrections which involve verb Three of

forms (4:15; 6:2, 58; 7:37; 8:54; 15:7; 20:18).

these involve synonyms (6:2; 7:14; 20:18) in which the scribe has changed from a less common to a more common Johannine form. Two others involve changes from a present to an aorlst subjunctive (8:54; 15:7). Both of these also conform to tendencies found in the original text itself. The opposite occurs at 4:15; but this corrects a reading shared only with D In a section where P66 is decidedly Neutral, and it Is clearly a change in the interest of making better sense of the text. In verb forms, therefore,

187
the scribe also tends to correct In conformity with, tendencies In the original text. 5* Although the scribe has a good record In the

original text (with respect to his basic tradition) in reading without the article before proper names, he adds the article eight times, when making corrections (1:46; 8:25; 11:35; 12:7, 16; 13:21, 23; 18:40). unls is a second

ary procedure as far as the basic text of the papyrus is concerned, and, if our above analysis is correct, probably so in terms of the Johannine original. - There is also a secondary harmonization involving articles in 12:9 and 12, where he has changed and both to read . We may conclude then that in matters of Johannine style the scribe corrects in a manner comparable to the "tendencies" of the original text of the MS. Although he sometimes changes to what appear to be original readings (e.g., 10:22 rorc 1. 6e), such corrections are probably for stylistic reasons; but most of the corrections appear to be toward a smoother, fuller text. On points of style, therefore, there is no indica tion that this Egyptian (perhaps Alexandrian) scribeturned-re censor is either creating a "scholarly" text or seeking to preserve the "Johannine original;" he Is rather producing a good, readable text. At the more significant points of variation,

188 however, where the change has less to do with style and more to do with the meaning of the text, there Is more ambiguity as to the principle of editorializing. Some readings are clearly secondary, and appear to be made in the Interest of an "easier" text. Kile Is particularly true of the three major changes toward the Western MSS. At 4:51 the change to eliminates the non-Johannine (its only occurrence in the Gospel).' Of the change in 11:33, Barrett has correctly observed: "The reading . . . is an easier text, since it avoids the statement that Jesus , and should be rejected as an editorial improvement* made in reverence for the person of Jesus."77 At 12:47 the elimination of the nega tive probably is in the interest of a sharp contrast between verses 47 and 48. Probably secondary, but not necessarily "easier," are the singular verb epetvev at 2:12, the plural at 4:25, the change to 6terpt3cv at 11:54 and to errtneaajv at 13:25, and the reading of the optative at 13:24. There are several other corrections which, if not

The correction also involves a change from indi rect discourse to a on-recltatlvua. This kind of change also occurs at 8:54. Although the o n -recitatlvua is more common in John than elsewhere in the New Testament (except Mark), and the scribe apparently prefers it, it is not possible on internal grounds to decide the original reading at such places. 77The Gospel According to St John, p. 333.

189
necessarily secondary, are probably in the Interest of preserving an "easier" or full text (1:19 add ; 2:11 om. ; 5:6 add r\bi\; 1:12 add ; 7:37 add pe; 8:21 add ; 9 J17 1. ; 9:18 add ; 9:36 om. e<pn; 12:3 add &; 14:11 1. * ;4:4 add ; 14:17 eoTat 1. OTtv; 17:14 add ex tpt; 19:15 e 1. eXerov). There are, however, five correctionsall of them deletionswhere the change is from a clearly secondary to a clearly original reading; and in each instance the cor rected reading is decidedly the more difficult (7:39 om. aYtov; 7:^0 om. ; 7:46 om. ; 10:26 om. etnov uptv; 14:4 om. . . 6) Here surely the scribe is after the "best" text in the sense of the "original" text, and not simply after a good, readable text. And it may well be that this principle also

underlies other changes as well.7 The difficulty, therefore, of final Judgment about

T^Out of 43 corrections judged to be "significant," P66c reads with Nestle-Aland 27 times. Most of these 27 are probably original to John; but they are not necessarily always the more difficult reading in terms of P66*. For example, the omission of in 2:11 is probably in favor of the original text, but since it also eliminates a ore'difficult reading (probably an early corruption due to a misunderstandingor right understandingof ), one cannot here judge whether the scribe is after the "origi nal" text or the "easiest" text.

190 the "principle of editing" la obvious. If the correction

at 11:33 ia patently secondary, the deletion of after inrcupa in 7*39 ie clearly original. factor remaine forever unknovm: choose not to change his text? Moreover, one major

How often did the scribe For if, in chapter 7, he

has eliminated the secondary readings at w . 39* 40, and 46, he has left others which are Just as surely secondary (v. 27 1. cpxnrat; v. 31 position of noXXot; v. 32 add oc; v. 36 , ); and in chapter 14 he has eliminated the secondary xat . . . oibare in v. 4, but has kept the

secondary for in v. 17 and added the secondary In v. 14. But one cannot evaluate him on what he did not do, since neither (or none) of his Vorlap;en may have pre served "original" readings which are to be found in other KSS. But over all the pattern of the corrections seems to be much like the tendencies found in the text of the original MS. Just as the original text preserves a major ity of original readings but tends to pick up a number of secondary readings, so also the scribe sometimes corrects in favor of the "original" text but also tends to pick up a number of secondary readings. This is particularly true in matters of style. We may conclude, therefore, that neither the original text nor the corrected text of P66 are in the direction of a "scholarly" recension. Just as the scribe is

191 careless in his original copying, even so he corrects without apparent controls. He gives no indication of having been a "textual critic" of the New Testament. The fact that he has copied the Gospel of John at all probably Indicates that it was "sacred," but that he copied and corrected in the fashion that he does probably indicates further that it was not "canonical Scripture."

CHAPTER V THE TEXTUAL RELATIONSHIPS OP P75 IN THE GOSPEL OP LUKE In contrast to its contemporary P66, the more recently discovered P75 has been the object of compara tively little investigation. The reason for this is not

bard to find. The work of C. L. Porter on its text in John, both in his dissertation1 and published study,2 gave clear indication as to the nature of its text. Whereas P66, as Birdsall correctly observed, is no "unsullied ancestor of Codex Vaticanus," P75 is not only in the ancestry of that great uncial but is so in a remarkably close way. The relationship between P75 and in John is far closer than the relationship between any two other New Testament MSS of any kind in the first four Christian cen turies. Moreover, this relationship is so close as to affirm beyond question the much-beleaguered opinion of Hort

lw

Textual Analysis," pp. 111-120. Papyrua Bodmer XV," pp. 368-376. 192

2,,

193
that "B must be regarded as having preserved . . . a very ancient text."3 Whether this text is, as Hort farther Maintained, "a very pure line of very ancient text" or not, Porter haa shown that represents a very pure line of descent of a very ancient text. If the Heutral texttype in John is a "recension," In the second century. j The text of P75 in John, therefore, lacks the it is a recension which was created

ambiguity of that of P66, and any additional studies must probably be in terms of the Implications of the conclusions of Porter. It is the significance of such implications, namely the role of P75 for KT textual criticism, which constitutes the chief interest in this papyrus to the purposes of the present study. Before that role may be assessed, however, an analy sis of the text of P75 in Luke similar to that of Porter's work in John is needed. Apart from general notices, which took into account its text in both John and Luke,5 only two

^The Kew Testament in the Original Greek. II, 250-251. The discussion of the problem of be found infra, pp. 252 ff. recension will

^See Floyd V. Pilson, "More Bodmer Papyri," BA, XXV* (1962), 50-57; P.-H. Menoud, "Papyrus Bodmer XIY-XV et 711 " RevTheoPhll, H I (1962), 107-116; Bruce . Metzger. "The Bodmer Papyrus of Luke and John," ExpT, . (1962), 201-203; J. de Savignac, "Les Papyrus Bodmer H V et XV," Scriptorium, XVII (1965), 50-55.

194
studies have thus far been directed toward Its text In Luke, and both of these were admittedly of a preliminary nature. All of these studies merely affirmed the tenta

tive conclusions of the editors: "D'une maniere genrale, est rarement du c&te de D la ou ce is. est seal; it semble avoir surtout del'affinite avec B."7 The present chapter, therefore, offers a more com plete analysis of P75 In Luke, using the method worked out in Chapter II. I. F75 AST) LUKE 10 P75 contains portions of the text of Luke from chap ters 3:19 to 18:17 and 22:5 to 24:53. Of this text 8:5-28; 10:8-17:29; and 22:38-24:53 are preserved almost in their entirety. About two-thirds of the text has been preserved

in most of the remaining portions, except for 3:18-4:42; 5:11-6:15; and 7:35-8:5, which are quite fragmentary. The

number of lines per page and letters per line remain con sistent throughout the papyrus, so that by comparing the number of full and partial pages preserved, one may Judge that P75 contains approximately 45 percent more text of

"Joseph A. Fitzmyer, "Papyrus BodmerXIV: Some Features of Our Oldest Text of Luke," CBQ, XXIY (1962), 170-179; Merchant A. King, "Notes on the Bodmer Manuscript of Luke," BibSac, CXXII (1965), 234-240. ^Martin and Kasser, Papyrus Bodmer , p. 29.

195'
Luke than of John. Although the present investigation is ultimately concerned with the entire text of P75 in Luke, for the purposes of giving direction to the total study, a complete analysis based on our method is first offered for Luke 10. This chapter has been chosen for very practical reasons: It is the first chapter where P75 reads almost without lacunae; it is one of the few chapters in which P45 con tains a large amount of text; and it is one of the few chapters in which C is complete. The total number of disagreements between the early KSS in Luke 10 is found in Table VIII. It is clear from these figures that the highest level of textual relation ship exists between P75 and B. A and TR. The next level is between

The third level is between P75 and K, and However, because these figures include

A V TR and C.

singular readings, and because the witness of P45 is not clear due to the incompleteness of its text, the table of agreements (Table IX) more adequately presents, and fully confirms, the relationships indicated in Table VTII. This table, which again is based on the double principle of "examining total variation" and "discriminating before. counting," indicates decisively that an extremely close relationship exists between P75 and B. Although these two sets of statistics are evidence enough of this very close relationship, the following

196

TABUS V I I I HOMBER OF DISAGREEMENTS BETWEEN EARLY MSS IN LUKE 1 0

TR TR
-

P45
48
-

P75
60

56 43
32

A 22

C 31

66
43
12
-

96 57 94 96 87
103

23 48 68
72

P45 P75
A C V

48
60

37
-

56 64 68 57
a

56 49 53 46 33
-

37 43 43 56 56 57 48

66
56
22 31

12

35
-

32 64 49 94 68

35 68 53 96 72

59
25

57 46 87 59

33
103
25

98
-

35
103

96 23

98 35

103

197

TABUS IX PEBCHiTAGB CF AEEMEJJTS IN LUKE 1 0 AT 7 0 UNHS CF VARIATION

TR TB
Pl*5

Pl*5 39.6

P75 2U.3 61*.l

20.0

It 25.6 51.0 61.6 67.6

A 82.9 28.3 2U.3 ZL.l*

C 71.1* 32.1 1*2.9 itl.U

V 80.0 32.1 15.7 12.9 29.1* 82.9

35.7 56.6
38.6 1*0.0

39.6 2l*.3 20.0 25.6 82.9 71.1*

5U.7
92.9

rft
A C D V

61*.l

5U.7
$1.0 28.3 32.1 $6.6 32.1

92.9 61.6 2i*.3 1*2.9 33.6 35.7

67.6 21 J* 1*1.1* . 12.9

35.3

1*8.5

51.5
3l*.3 3l*.3

35.3 liB.5

7U.3

7U.3 3l*.3 82.9

65.7
32.9

35.7
80.0

51.5
29.1*

3U.3 65.7

32.9

NOTE: ?h$ reads at 53 places; at 68.

198 observations indicate the full extent of that relationship. 1. Of the twelve disagreements between P75 and in this chapter, six of them are singular (or nearly so) to P75 or B: 1019 add post 10:24 P75 U 998 syC 10:27 * oa.

on, post om. ante TIJ O 6

10:31 1 579 10:37 * 10:39

oa. o ante

75 <

There is one further reading where P75 and each has a nearly singular reading against the rest: 10:18 P75 472 254 579 A C D V rell ex ex ex

It will be observed that none of these is a textually significant variant, and except perhaps for the omis sion of ev at 10:31* none of them has a prima facie claim to represent the original text. There are five other disagreements between P75 and where each has a reading supported by one or more of the early MSS: 10:39 P75 P45 H E 579 B* A C Byz TR xat n *<u

D
10:39

P75 P45 A W Byz b TO 1 * D L 579 892 itP bo syc xaTeXctnev

10:40 P75 P45 D Byz TR A C L pm

199 J
10:42 P75 P45 i C V Byz TR trot 6c coftv xpcia 6c xpeta eortv 1 33 579 C o\tT*v 6c cortv xpeto * 6c cortv t j 38 syP*l 6c cortv xpcta 10:42 75 P*5 T l d A C V Byz TO D L 579 ; Of these variants only the first reading In 10:42 Is textually significant,0 and at this point et al. not only preserve the Keutral reading, but most likely the original as well.9 also preserves the Neutral text at 10:39 () and 10:42 (om. an ), while P75 preserves the Neutral reading at 10:39 (*, ) . It nay be of Interest to note further that P75 and read together without any variation whatever for the first eighteen verses, covering two distinct pericopes, and that the five non-singular disagreements all occur in a single perlcope.

10:39 / may be so, but each is abbrevi ated in the KSS to and ", which makes it an easy place for error. *At this point the Judgment of Eort seems better than that of the translators of the NE3. See Tasker, ed., Greek New Testaaent, p. 420. The translators of the NEB considered the reading of et al. to be a conflation of the reading of P75 et al. and of 3 3 syPal. It seems to be asking too much of the thirteenth century codex 38 (not a necessarily good MS) and the poorly preserved syPal to suggest that they alone preserve an early second century reading (J). Their text is much easier to explain as a later (and more logical) shortening of the text of B. The choice therefore is between the "shorter" reading of P75 et al. and the "more difficult" reading of et al. Since the shorter reading fits the context so well, one is hard pressed to find a reason for its having been changed into the longer.

200 In order to indicate how minimal the amount of dis agreement between P75 and actually is, a comparison of the disagreements in Luke 10 between some of the MSS in Family 1 may be noted. According to Kirsopp lake the closest textual relationship in this family exists between Codices 118 and 209. b relationship is so close in fact Tthat lake could argue: "Either 118 209 have a common Tbe next level of

archetype X, or 118 is a copy of 209. " 1 0

relationship in this family is between 1 and 209. In chap ter 10 of Luke, 118 and 209 have eight disagreements, none of which is a singular reading to either codex, and 1 and 209 have fourteen disagreements. This means that in terms of disagreements, P75 and in Luke 10 are within the same range of relationship as exists within the "inner circle" of Family 1, 2. The closeness of this relationship is further

demonstrated by the agreements between P75 and B. There are four variants where P75 and are the only uncials to support one of the readings: 10:1 10:6 P75 579 700 713 e U C D V L H N rell antaretXer exei exct exct n

P75 433 1012 itP* Or A C V L Byz c e sy l604 pc TR

Codex 1 of the Gospels and its Allies (Cambridge, 1902), p. xxiii.

201 10x21 P75 aur vg nveuuaTt * D 1241 a b 1 ev ayta C 1 2 L 33 c e ff r* ev tnreuuaTt art 579 mreuuaTt nveuuaTt A W Byz TR 45 ev 10:27 75 A C V D rell There are six other readings where P75 and are Joined by only one or two of the Neutral uncials; 10:5 P75 A W D a C L 10:6 P75 rell 1241 e Byz sy8 bo 1 3 otxtav otxtav otxtav aur c f vg otxtav * 579

10:15 75 L 0115 157 726 1375 bo rell 10:24 75 0124 sa rell t

10:27 75 0124 472 A C W Byz aur e f vg D a b c ff 2 1 1 q r 1 10:30 P75 * C 8 rell

and four others where they are supported by D or P45 against the rest: 10:2 10:15 P75 D 0l8l 700 e rell P75 D 579 syS-c rell 5

10:35 P75 45 sa Dee rell

202

P75 45 ea omit tt* C 1 33 579 * otxtov A D V Byz i t vg TR They further agree in eleven Neutral readings against D and the Byzantine MSS (10:3 om. >; 10:19 btbwxa 1. ; 10:20 1. ; 10:21 eufioxta
CYCVCTO CYCVCTO

10t38

1.

euftoxtoj 10:25 om. xai 2; .10:32 . . . c\ev 1. 10:39 <> 1. trapa;

. . . [P45 D] or . . . cXearv [A C V Byz TOlj 10:36 om. ; 10:38 om.

CYCVCTO;

10:39 vapaxaQcaBttaa 1. napaxaOtoaoo; 10:4l . ), as Hell as two others where D Joins the Neutral KSS against and the rest (10:1 ClO:173 add ; 10:35 . ).
Although matters of orthography have ordinarily been left out of the discussion of variants, it may be noted in conclusion that P75 and agree with very few others at four such places in this chapter: 10:7 10:13 P75 X 579 > rell ctovc

P75 A D 579 a C L R X Byz P45 U P75 45 rell cov

10:35 10:42

75 3 1 579 . rell uapta

It is clear, therefore, that P75 and have at least as close a relationship in Luke 10 as they do in John. Moreover, this relationship is of such nature that one may

203 safely conclude that the text of In Luke existed In most of its particulars at the end of the second century. If

and have been considered in the past as representing a texttype, then P75 and appear to have a family relation ship within that texttype. It remains only to he shown

that this same relationship is maintained throughout the Gospel. II. P75 AND IN THE REMAINDER OP LUKE

The percentage of agreements between P75 and the other early MSS on a chapter by chapter basis is presented in Table X. It is at once clear that the relationships *

found in chapter 10 are constant throughout the Gospel. It say be noted further that in many chapters P45 and C are missing. Were one to add any two other MSS in these chap ters to form a broader base for tabulating variants, the percentages of agreement between P75 and would be even higher than it is. Perhaps the best demonstration of this relationship is to note the total number of disagreements between the two MSS, such as Porter presented for John.11 list of such disagreements totals 205. ^ Porter's

The list of

llw

Papyrus Bodmer XV," p. 368-374.

^This figure should be raised to 208. Porter has left out the following: 3:25 the singular reading of

204
TABLE X CHAPTER BT CHAPTER PERCENTAGES <F AOEEMENTS IN LUKE BETWEHI P75 AID CTHSR EARLT IBS

Chap No. of Chits

TR
29.6 16.1* 303 21.5

P2i5
t

85.2 96.1* 97.0 95.3 88.2* 92.9 86.7 87.3 82.0 81.8 88.2 93.1 88.6 96.9 87.8 89.3 90.7

C
29.6 17.1 22.2 29.0 2*6.1* 1*2.9 39.0

D $$.6
29.1 15.1 26.2

>5
6 7 8

27

lac 71.U

59.3 ' 33.9

55.6 2*5.5
57^6 15.1* 37.6 15.7 38.7 25.1* 36.0 11.2* H*.7 17.2

55
33

63.6
63.6

20.0 2l*.2 21.5 31.9 2l*.3 1*1.3 29.5 1*2.0 13.6 29.U 20.7 13.6 17.2 20.3 17.9 25.1*

lac lac
6$.Q 61*.l

65
69 70

58.5
66.7 67.6 62.7 59.2

9
10 11 12 13 1 2 15 16
;17-18

A.8
39.6 38.7 33.8 2*0.0

33.3
38.6 32.0 26.8 32.0

75
71 50
2*2*

5U.7
1*8.0 50.0 70.8

lac lac lac lac lac lac lac

53.2

6k.O
68.2 67.6 72.1* 61.1* 70.3 70.3 67.9 65.2

9.1
17.6 20.7 13.6 12.5 12.2 16.1 21.8

36.U
35.3 I*lt.8 1*3.2 3l*.l* 21*.3 35.7 32.7

32* 29
1*2*

lac lac lac lac lac lac

9.1
17.2 13.5 12.5 25.1*

22 23 21*
TOTAL

62* 71* $6 860

lac
1*0.7

58.5

NOTE: Pl*5 shares readings at 253 places; C at 396.

205 disagreements in Luke, which will be found in Appendix III, totals 229 When one takes into consideration the amount

of text preserved in each of the Gospels, the number of disagreements are considerably fewer in Luke than in John. Whereas in John there are approximately 5*5 disagreements per page of KS, there are about 4.4 per page in Luke. In percentages this means that while P75 has 45 percent more text of Luke than of John, it has only 9.6 percent more disagreements with in Luke than in John. Tables XI, XII, and XIII list the 229 disagreements between P75 and according to the more frequently occur ring types of variation. Tables XI and XII show the dis

agreements which are the result of singular or sub-singular readings in P75 and respectively. Table XIII presents a

similar list where P75 and are Joined by one or more of the important uncials. It will be noted that well over one-half (57.5/0 of the disagreements are singular and sub-singular readings in the two MSS, which probably means that most of the variations are due to nothing more than the habits of the two scribes involved. But more significant is the fact that of the 229 disagreements, only 18 are of such nature as to make a difference in meaning in the passage in which

TWV before in B; 6:24 P75 ] ; 8:31 P75 pcvtire ] petvtjTe.

206
R C M A M

ST
*

C C * *** J

et * et rs *e cc - * ^ 3 4 ^- -
i

-* ^ CJ

* *
et *->

> * > >

. h

* r

"W

et o >r

- 11:88 - 15:82

* !
0 V S tI "8.

. t.

P. t* 0 -

11:10 14:38

fl S .

e n t * ^ ~4

t 9 ft. O

m 9

e .

ru a ft.

ft.

S . U a -

^< ^>

.
>

- c* et

CD

^ > ? 4" * -*-* et

**

* CD t"HC4-!

2 *S
C - ~ %4 * 3 "

*>

fat

* ^

** ci
-4 -4

CM

a o ** rt h hJ3 < -"-J 10 . C m O W E ft, C ft.

> * ** - C . 1-4
"4

fa

et e

ci

fi3

CO

a>4 . * <- C4 C4 t - C4

a + > co

<

- ) - 1-4 -

00 < 0 CM I *> CM CM CJ

-r'^caoics^cii-eoe\jeM^ 7~9^
_I,-I,*_l.-Ir-If-I*-C1CM J > t - t - > > t l l l l l l l

ocVontoi
r< r4 >-) P4C4

t +

Ti??^t?5?<?
et et
I - C

? ^ * ? ? ^ * *? 5 P V
e > * H o n io ci r-cMi-tcicsTfioooeMrt-*

c:r*cinH

BOlOlMr4i-ll

co e7 ci ci c7 e7 >' -r cV
r t r t r t H H H H M C I

ft.

207
*
0
* - ^ ^4 - * e t

315
-

II

fa C

<^<-<Fi-'e,^f5

..........
I I * > I

-* cs o ca
H H f 4 t l

iJII
t

>

>

C O

*-4 *

O C5 X

c*fa
4

ctnco e o a
1-l - * m*

>

fa e fa

~ *

-*
en K 4 e .

a. a > a

"2
n o en oi es I t-i we
0 C 4

fa fa

-4

- 4

-a CD c r- o o o a I + I I *

-r*
>=

- <

> > t**>

C C

" e fa _ 0

J3
> fa

*>

"5

fa

-*

- 3 C <-< = - JS -* t * fa m -r* * 5 **> J C * H

c o

-< >

fa >
-

- *

TO fa rf> CO

a
-

fa

3 *

8.3
C -

6 m > ,

f 5 -

+ > co .

S
0

C* "9

20
* 9 r> < * t !

*
** . h C C * v . l. -* s >

O ^ O f J N H O C f i H * ^ * >* * * Q 0 * * * * * * - * >

* /

>

* -

>

1
net*
-< ^ TJ 3

1 f> '

>

CD *

* t

f o o r
* 4 1-4

25
(* *

>

*> ** C .

0000003->-

, - l i-H <-l r-H f t rt C

* i. .*

>* tj

. w V C * . > > * e
>

3 C Ih S. a.

<v a -*

en

-<

. fa

<

tor-col-

* c + * c -W
* Q * IB * J3 *>

_ . . . . . . . . . . . . _ _ _

cu a, e, . 03

"*
*

*
><

fa

H H H H I H H I H N C I C I M

+ + +

+f-f-

8
+

I ^ f 00 i * - cMtMci' t>ncoiO'-<2

O N r t N C J C J V N n ' o n H H H H r t W N N C J + I t t + I + I + I

s
c e u

(. >

l >

> + > > > ! > ?

>oeo O O N H N n r - o o N M O M r t W O M C I M M f i f C O l O O e e o o o o o H H c i o i r e e i i - n r

209 they occurand 11 of these are eingular readings. This means that the following seven variants are the only places in the Gospel of Luke where P75 and have s i g n i f i c a n t l y different readings with M support: S 10:42
1

11:25
11:33 11:48 12:39

P75 P45 A C* W Byz TR 6c xpeta 6* xpeta L 1 33 579 Cr eortv xpeta * oXtrr eortv ai 38 s y P xpeto 75 * A D W Byz TR omit BCLTRTXqppc
P75 P45 L pc Bit A C W Byz TR 75 A C D W Byz TR L 892 1241 75 * D e i syS 0 W L pa r A Q 070 Byz TR omit xat omit ptodtav

13:33 f 15:21

75 D 69 157 1241 pc V L Byz TR 75 A W L Byz TR D X U 1 33 700 1241 pc

I t i s important to note further that of the 96 read ings where P75 and disagree in conjunction with other MSS, such disagreement i s rarely in favor of the singular or sub-singular agreement of e i t h e r with D. P75 reads with D 72 983 a b eye at 14:23 (where the scribe has corrected by d e l e t i n g ) ; adds ante with D and 69 at 14:35; and omits with D i t s y s ' c at 23:39. with D at 15:4. reads for ^ with D pc at 13:28 and for Only the omission of at

210 23:39 oay be etyled "Western'1 In any sense of that word.!3 It is clear, therefore, that not only do P75 and have a very close relationship In Luke, but that any varia tion between them is not the result of a "mixture" of readings from another textual tradition. There is one further item which sets forth the close ness of the relationship between these two MSS throughout Luke, and that is the number of instances where they read alone (or nearly so) against all other MSS. In his "indictment" of the Neutral tradition, H. C. HosJdLer includes a "rough list of approximate solecisms" of in Luke.1* This list contains 93 readings, of which P75 now shares a reading at 65 places. Of these 65 singular

readings, the text of has now been found to exist in the second century (in P75) at 3* places. These include the following textually significant readings: are missing in Hoskier's "rough llst.")^ 6:26 6:31 om. ot om. : (Starred items

*3p75 may also have singular agreement with D at 3:36; 6:22; and 9:27. But there are lacunae in P75 at these places, and such agreement is only conjectural. See the discussion at the end of Appendix III, pp. 295 ff. ^Codex and Its Allies, I, 208-216. !5it should be noted that other discoveries since Hoskier, notably P45, often agree with P75 and in these lists. I am simply using Hoskier's list as a convenient starting point.

211

*8:l6
8:25 8:45 9:62

Offl.

tva ot ctonopcuopevot 0Venotv

om. om. at om.

(cum P45)

10:24 add post axouoat 10:38 11:11 11:11 11:12 17:23 24:37

om. ex ^ (),

(cum P45) (cum P45)

om. xat xat 1. om.

(cum P45) (cum P45 L)

om. aneXSnTC
l.

Although the remainder of the singular agreements is less significant in terms of the meaning of the text, the very frequency of such agreements at these less important places indicates the extremely close relationship which does exist between these two MSS. 5:3 6:26 6:34 6:37 6:37 6:46 8:5 *8:6 8:23 8:35 I note the following:

ex 1. e6t6aoxev ex etneotv 1. etncwjtv om. (cum P45)

6txacTe 1. 1. 1. 1. om. ante 1. om. ante

212 10:1 on

10;27 om. xot 1 12:22 12:56 14:1 14:12 *l6:3 16:6 on. T O Y xatpov 6c 1. T O Y 6 C xatpoY on. T Y ante pij 1. pqoe add post
t

, toxvm

ypafoY > 1. |

17:12 on. post 17:28 otxooopouv 1. 23:7 23:19 add T O Y ante . 3 (l)

(cum )

23:49 add at ante 24:24 on. xat 3 24:28 24:38 (D It also omit In a different order)

1. on. 6ta

Singular agreement In minutiae such as these would seem to Indicate conclusively that P75 and not only rep resent a texttype, but do so as the closest of "relatives." The discovery of P75 now makes it certain that the text of existed in the second century both in its main features and in a great many of its particulars. Finally, it remains to be asked whether either of the MSS appears to reflect textual or scribal characteris tics of any kind, which they do not have in common.

213 III. TEXTUAL CHARACTERISTICS OP P75 AND

It was noted above that over one-half the disagreeaents between P75 and are singular and sub-singular readings in the two KSS. Tables II and XII indicate that there are fewer of these in than in P75,*6 and a careful look at Table XII reveils no pattern "or direction of "editorializing" in B. The single *significante omission C11:42 om. 0eou) is probably a scribal error, and such harmonizations (with Matt. 26:17) as oot at 22:9 are few (cf. 11:25 and 15:21). Often what appear as tendencies in B, also appear in 75 B, for example, tends to omit the preposition in partitive phrases (10:52; 12:58), but so also does P75 (17:7). Hoskier accused of reading simple words for com pound forms, 17 which it does singularly at 6:38 and 15:24. But in the four examples Hoskier used, P75 also reads the simple form, and does so singularly at 11:22, 13:25, and 15:22. has five instances of singular omission of the article, and P75 has ten. Moreover, they have three instances of singular agreement in omitting the article (dz6, 35; l4:l) and many other places where they read with

the Neutral tradition against the rest in so doing (e.g.,

^This list for B, of course, does not include all of Luke, but only the points where P75 also has a reading.
17

Codex and Its Allies, I, 248-249.

214 5:3 5 bisj 6:30; 8:41). What this means, therefore, is that alaost any "edi torial" tendencies found in are anticipated by its closest relative and are probably only a reflection of close adher ence to the Neutral tradition. This, plus the fact that the number of singular readings in is Bdnlxal, means that the scribe of this KS is alaost totally free from "editing" his text. He is first of all a copyist, and apparently one of high integrity with respect to his exesplar.lS Sable XI, on the other hand, indicates that P75 has one clear tendency, namely to read a "shorter text." This is especially true in regard to pronouns and conjunctions, but it is also true at more significant points. The scribe adds a pronoun only at 11:7 and 12:18 (both possessives) and in each instance has corrected his text. Ee adds a conjunction twice (9:28; 9:48), and there is only one significant singular addition in the entire text of Luke (l6:19 add ), a reading which, because of its Interest, has already received special attention.^

only other alternative is that the scribe of had more than one MS from which he copied, and that he "edited" by choosing to stay with a text like P75^see especially Henry J. Cadbury, "A Proper Name for Dives," JBL, LXXXI (1962). 399-402. Idea, "The Name for Dives," JBL, LXXXIV (1965), 72. The =encion of this particular reading in almost every overview of P75 tends to distort the true textual nature of the papyrus, which has no other such singular additions. (The long "correction"

215 It will be noted further from Table XIII, that where P75 and disagree with other MS support, the tendencies in P75 to read the shorter text often continue. This is true both of pronouns (8:20; 11:22; 12:22; 17:6; 24:39) and of the more significant variants (11:25, 33; 12:39; 15:21). Although this may merely reflect the scribe's habits, an examination of the Neutral text in Luke indicates that P75 is closely adhering to "Neutral tendencies." There are, for example, 26 places in our total units of variation in Luke, where P75 and agree against the early Byzantine MSS (A C V and sometimes D or ) and the TR in omitting the possessive pronoun or a direct or indirect object.20 There

are at least equal that number where the early Byzantine KSS join the Neutral tradition and D against most of the rest in such omissions. But far more important are the following examples of "short text," which do not Include numerous other instances involving a single word: (Other

KSS supporting the text of P75 are in parentheses.) 5:38 6:26 6:45 om. xat om. ot ( D L pc)

(700 1241 sy3 sa) ( D L V pc)

om. 2

at 17:14, borrowed from Matt. 8:3* is clearly the work of a later hand and does not qualify this statement.) 2 0 5 : 5 ; 6:26, 40, 45; 7:6; 8:27, 49; 9:51, 54, 62; 10:1, 33, 35; 12:22 bis, 41, 53; 13:27; 16:1; 17:11, 12, 33; 22:49, 55, 57; 251.

216 8:l6 8:25 8:43 8:45 8:45 8:54 9:54 om. tvo ot etonopeuouevot om. (700 aeth) ([D] sa sy8) (1574)

om. 0tov om. at ot ()

( 700 pc sa sy 8 0 ) ( L 22 157 Pc)

om. xat , om. xat om. xat

( D L pc itP1) (45 L 157 544 579

700 990 1241 sa aur e 1 vg sy 3 0 ) 9:55-56 10:22 om. xat . . . (p45 A C W L pm) (45 D L

6m. xat

33 579 892 1241 pc aur a b e f sy 8 0 ) 10:38 11:2 11:2 om. otxov (45 sa) ( L 1 22 aur sy 5 )

om. & ev

om. , ev xat cnt ( L ff2 sy 5 0 ) ( L 700 sa

11:4

om. vg sy5)

11:11

om. , xat ff2 1 sa sy 8 )

(45 440 i

11:44 om. xat 892 1241 a c e ff2 1 vg) 11:48 om.

(P45 C L 33

( D L 579 124l sy8 a b e 1 r 1 ) (45 L 579 892 pc sy3)

11:54 om. tva 17:23 17:24 22:31 om. ()

om. om.

(D 220 a b c e l ff2 sa) (L 124l sa bo sy 8 )

217 22:43-44 22:64 22:68 om. om. . . .

CTUTTTOV

(69 A W R pc ea bo f ey) ( L b bo)

om. \ ( L 0153 157 bo)

23:17 om. 6e ctxcv (A L 0124 124l a sa bo)

23:23 om. xat ^ '<

i

( L 130 1241 itPl) ( c c D W 0124 ft

23:34 om. . . . 157 545 pc sa a sy8)

.

24:1

om.

( C L 33 579 124l bo it) ( D W L 579 e sy8)

24:42 om. xat 24:46 om. xat e6et

(K C* D L 579 sa bo It)

Indeed, there Is not a single Important Instance of "short text" In which is not now supported by P75 (i.e., where P75 has text). Die fact that P75 has the four additional instances of "short text" against is perhaps an indica tion that P75 is the truer witness to the Neutral text at these points as well. Moreover, the fact that P75 and are often Joined by D and the OL, as well as by others, and that in the great majority of these the "short text" is the preferred reading on all grounds of internal criticism (many are harmonizations with Matthew and/or Mark), seems to indicate further that this is not simply a "Neutral" tendency, but a reflection of "Lucan" tendencies as well.2*
01

Although it is not our concern at this point, the

218 This does not mean, of course, that any of the singular "omissions" of P75 necessarily has claim to represent the original text (although the omission of voootov at 13:3^ and at 14:8 [with b sa sy^S] certainly belong to the "Neutral" pattern); but this scribe's almost adamant stand against additions to his text does seem to indicate that he is attempting carefully to preserve the text he is copying. We may best conclude the investigation of tendencies in P75 by noting two recent studies which spoke to this question, and each of which has taken a quite different turn. In his 1965 presidential address before the Society of Biblical Literature, Kenneth W. Clark had occasion to examine some of the singular variations in P75. 2 2 conclusion was that in general, P75 tends to support our current critical text, and yet the papyrus vividly portrays a fluid state of the text at about A. D. 200. Such scribal freedom suggests that the gospel text was little more stable than an oral tradition and that we may be pursuing the retreating mirage of the "original text" (p. 15). However, Clark's choice of three variants from Luke His

fact that P75 is so faithful in reading a "short text" may be significant as far as the "Western non-interpolations" are concerned. For P75 does not read one of them. At, least this MS clearly puts the whole problem well back into the second century. "The Theological Relevance of Textual Variation in Current Criticism of the Greek New Testament," JBL, LXXXV ' (1966), 1-16. -~~ _ . .. .. .
22

219 to portray "vividly" the "fluid state of the text" is not a very happy one, Hie addition at 17:14 hae already been noted to be the work of a later hand. (Clark is quite

misleading at this point in saying that "the scribe of P75 borrows. . . .") Moreover, his "unique reading heretofore unreported" of for at 11:11 seems to be a "nonsense" reading pure and simple, since the is immedi ately followed by . Obis error is surely a form of "mental metathesis" and not a case of "scribal freedom." The other reading, for ^ at 11:24,

is not singular to P75 as Clark suggests. It is also read by 1 579 1200 1375. But in spite of later support, this reading looks like a case of error, rather than of scribal freedom. The exact thing occurs in P75 at 14:8, and here

the accompanying makes the reading impossible. One eight compare also 9:32 1. ; 11:53 . ; 17:4 1. ; and 23:29 1. , all of which are uncorrected and impossible Greek. In contrast to Clark, E. C. Colwell had previously concluded his examination of the scribal habits of P75 with this Judgment: In P75 the text that is produced can be explained in all its variants as a result of a single force, namely the disciplined scribe who writes with the intention of being careful and accurate. There is no evidence of revision of his work by anyone else, or in fact of any real revision.

220

. . any impulse to iaprove style is for the most part defeated by the obligation to sake an exact copy.23 Our present examination seems fully to confirm this judgment. Rather than reflecting a fluid state of text, as P66 in John surely does, P75 in Luke appears to reflect, as does B, a careful copyist preserving his original text. Any "editorial" tendencies in P75 are those already found
1

in his textual tradition. 17. CONCLUSIONS From the analysis by Porter of P75 in John, and from the foregoing analysis of its text in Luke, the folloiring important conclusions should be noted. 1. In contrast to the text of P66, which is mixed" both in terras of textual traditions and original and secondary readings, the text of P75 is almost totally "pure" in terms of its textual tradition, and it appears to be much more reliable in terms of the original text. Whereas none of the singular and sub-singular readings of P66, and few of its readings where it varies from the Neutral tradition, seem to preserve the original text, there are a number of readings in P75 where it varies from B, as well as some of its singular readings to which one

"Scribal Habits," pp. 381 and 386.

221 m e t give serious consideration in terms of original text. This is especially true of the omissions at Luke 11:33 (ouoc uno ), which it shares with P45 L E T 0124 pc, at Luke 12:39 (cypnYOpnoer or xat) which it shares with R* D e i sy 8 0 , and at John 9:3-39 ( 6 e*n niffrcus, xuptc xat irpooexuvnoev . etncv ), which it shares with K l b (l) bo Q . i 2 Again in contrast to P66, the scribe of P75 appears to have taken few liberties with his text. Vhereas the scribe of F66 is both a careless worksan and a con scious "recensor," the scribe of P75 is a careful workaan and appears to be almost free froa "editorial" activity. His chief concern seems to have been the careful preserva tion of the text which he was copying. 3. Because of its extremely close relationship with B, it is no longer correct to speak of 3 as "recensional,B If by this term is meant either a recension in the third century or the scribe of as a recensor. The text of P75 froa the second century appears to absolve the scribe of of carefully "editing" his text. Rather he is carefully "copying" his text. 4. Finally, the text of P75 in Luke and John indicates that the Neutral texttype, as a fully developed texttype, existed In all of its essentials in the second century. Furthermore, it so existed across two distinct textual histories (Luke and John). If other texttypes are

222 processes which developed over centuries, or are "uncontrolled popular texts,"2* the Neutral texttype is both a completed "process" by the end of the second century, and apparently was one which was handed down in a carefully controlled tradition. Therefore, such descriptions as those of P66 and P75 by Werner 0. Kummel in hie latest revision of Feine-Beha's Introduction are quite misleading, if not in error. Of P66 he says: "The codex . . . presents a text which . . . be-

longs to a preliminary stage of the 'Egyptian text before the recensions."25 The text of the contemporary P75 seees Of P75 Kummel has

completely to nullify this Judgment. said:

"This text is a predecessor of the 'Egyptian' text-

form*1 (p. 364). Again, the conclusions of this chapter indicate that such terms as "predecessor" or "pre-recensional" are not permissible while speaking of P75 Any "recension" or development of a text-form precedes the text of P75.

This is Colwell's description of the Western texttype. See "Origin of Texttype3," p. 137. ^introduction to the New Testament, 14 th rev. ed., trans. A. J. Mattlll, Jr. (Nashville, 19bt>), p. 364.

CHAPTER VI THE SIGNIFICANCE OP P66 AND P75 FOR THE PROBLEM OF METHOD IN NEW TESTAMENT TEXTUAL CRITICISM It was noted in Chapter I that the paradox of contemporary criticism is that the recensional character of the Neutral texttype is considered an "assured result," while at the same time most contemporary critical texts have a predominantly Neutral appearance. Some critics have considered this paradox to mark the failure of contemporary eclecticism as a total method. In his 1947 discussion of method, E. C. Colwell concluded that "no patching will preserve the theory of Westcott and Hort;"1 and "patching" fairly describes how some have considered eclecticism.2 Hence there has been an appeal for a

new methodbeyond eclecticismto replace Westcott and Hort. Such Judgments as these, however, were based chiefly

ln 2

Genealogical Method," p. 132.

Cf. the opinion of Clark quoted in Chapter I, p. 6. 223

224 on the premise that the "recensional" nature of the Neutral texttype invalidated the textual theory of westcott and Hort ae well as that of their "offspring," that is, it invalidated any theory which considered a textual tradition to have preserved a "relatively pure line of very ancient text." These Judgments, however, were all made before the discovery and analysis of P66 and P75 The position taken

in this chapter is that an eclectic method which seeks a proper balance between external and internal evidence is in fact a valid method, and that the analyses offered in this study contribute to the validity of such a method in at least three ways. First, they contribute to the question of the relationship between internal and external evidence in eclecticism. The problem here is whether eclecticism should be a

thoroughgoing methodology after the manner of 0. D. Kilpatrick, or whether evaluations based on external considerations such as the date and general quality of a MS or MSS should also be a determining factor. Secondly, P66 and P75 contribute to the important question of early textual recension. It is the conclusion

of this study that the combined witness of P66 and P75 tend to undercut the first horn of the contemporary dilemma, namely the recensional character of the Neutral texttype. If this is true, then the major cause of unrest over

225 contemporary eclecticism will have been removedor at least the embarrassment over the Hortian face of our contemporary critical texts and translations will have been removed. Finally, if it can be demonstratedor at least if probability thus indicatesthat the. Neutral texttype is not recensional, then a new evaluation of the textual theory of Westcott and Hort may be in order. Instead of

viewing the original text as totally lost and scattered in random fashion throughout the MS tradition (which is what Kilpatrick's method seems to imply), one may perhaps be prepared to view the Neutral texttype as a "relatively pure" representative of the original. Such an evaluation of course depends upon the demonstration of the relative superiority of the best representatives of this tradition over other MSS and MS traditions. The chief difference between such a theory and that of Hort is that more emphasis will be placed on the term "relatively" than did Hort. The reason for this is that

the total application of eclecticism as a method will place more emphasis on Internal evidence, with the result that more often one will be willing to view the texttype as having been corrupted.

226
I. P66 AND P75 AND THE ROLE OP EXTERNAL EVIDENCE IN TEXTUAL CRITICISM

One of the basic problems of contemporary eclecticism ie that of finding a proper balance between external and internal evidence. How ouch weight does one place on such matters as the date and character of a KS or families of MSS, and how much weight on purely internal and contextual considerations? The term "eclecticism" has been used to describe two forms of answer to this question. On the one hand, it is

used of a method which tries to find the fine balance between internal and external considerations, a method in which "verbal criticism, external and internal criticism all have their part to play and must, give each other mutual support."^ This method has been called "reasoned eclecti-

cism," and for the most part describes what is probably the reigning contemporary method. On the other hand, the term "eclecticism" is also used of a method which relies chiefly on internal considerations. This is particularly true of the work of

^Vaganay, An Introduction, p. 91. ^That it was openly espoused by the translators of the NEB may be noted both in the "Introduction" and in the textual notes (Tasker, ed., The Greek New Testament).

227 C. H. Turner5 and 0. D. Kilpatrick.^ To thie method Xilpatrick has applied the term "rigorous eclecticism;"^
Q

and although he suggests that he does not intend "to make

l i g h t of the importance of knowledge of manuscripts* 0 in actual practice h i s method often does so.9 In many respects Kilpatrick's "rigorous eclecticism" stands In direct contradistinction to the method of Vestcott and Eort. For Hort there was a primary dictum for "Knowledge of documents

the evaluation of any variant:

should precede final Judgement upon readings." 1 0 He argued that Internal considerations are always secondary t o the evaluation of the documents themselves: "The uses of i f taken

internal evidence are subordinate and accessory:

as the primary guide, i t cannot but lead to extensive 5see e s p e c i a l l y "Marcan Usage: Notes, Critical and Exegetical, on the Second Gospel," JTS, XXV (1923/24), 377-336; XXVI (1924/25), 12-20, 145-1567^25-240, 337-346; XXVTI (1925/26), 58-62; XXVIII (1926/27), 9-30, 349-362; "A Textual Commentary of Mark 1," JTS, XXVIII (1926/27), 145-158: "Western Readings in the Second Half of S t . Kark's Gospel,* JTS, XXIX (1927/28), 1-16. See supra, p . 5, n. 10.
7"Western Text and Original Text in the Gospels and Acts," p. 36.
8

Ibld.

9This may be seen in his various articles on lexical and grammatical usages of NT authors, as well as in his contribution to the Casey Festschrift, "An Eclectic Study of the Text of Acts."
10

The New Testament, II, 31.

228

Kllpatrlck, In contrast t o t h i s , says of his own aethod: "Thus we do not concern ourselves with attempting t o s a t i s f y ourselves that the Egyptian [Neutral] text or the Western text as a whole i s right but we try to decide each variant by i t s e l f . " 1 2 d i e i s similar to the three dicta offered by Frederick C. Grant: .1. Ho one type of text i s i n f a l l i b l e , or t o be preferred by virtue of i t s generally superior authority. 2 . Each reading crust be examined on i t s merits, and preference must be given to those readings which are demonstrably in the s t y l e of the author under consideration. 3- Readings which explain other variants, but are not contrariwise themselves to be explained by the others, merit our p r e f e r e n c e . ^ What Kilpatrick and Grant c l e a r l y imply i s that the primary considerations are always internal. I t would

appear that the chief value of the MSS i s to supply the variants with which the c r i t i c i s t o do his work! Epp has cogently asked of Kilpatrick at t h i s point: Bldon J. *Why

not add a few more conjectural readings on the assumption that they may have been l o s t at some point in the history

ibid., I, 543.
An Eclectic Study," p. 64.

12,1

^"The Greek Text of the New Testament," An Introduction to the Revised Standard Version of the New Testaaent (International Council of Religious Education, 194b;, p. 41.

229 of the text?"14 As to the weakness of this procedure and therefore In partial defense of Hort, the foregoing analyses of P66 and P75 Indicate not only that MSS may be Judged as to their general quality, but that this Judgment must often be a factor In determining the preferred reading. This is

particularly true in three areas where "rigorous eclecticism" appears to be an inadequate method: (l) the evalua-

tion of a singular reading as the original reading because it "fits" an author's style, (2) the failure to reckon fully with the possibility that an author may vary his style, and (3) the problem of evaluating readings where Internal considerations lead to a stalemate. Singular Readings and Original Text. Chapters IV and V of this study have Indicated that the internal considerations for which Kilpatrick would argue as a basis for the recovery of the original text, may Justly be used first for the evaluation of existing MSS. It seems to this writer that the evaluation of the MS itself should be a primary consideration before one considers the reading of that MS at a specific variant. Kilpatrick, for example, often opts for a reading which has very little MS support, on the basis that such a

"Some Important Textual Studies," JBL, IXXXIV (1965), 174, a review of the Casey Festschrift in which Kilpatrick's "An Eclectic Study" appeared.

l4

230 reading best reflects the given author's style. But serious questions arise when his preferred readings are singular to such witnesses as D, 124l or syr1*.^ The read

ings in question are not necessarily condemned because they are singular, but because they are singular in MSS whose texts abound in singular readings of patently secondary character Our evaluation of P66 offers a case in point. In

the flurry of notices which followed the publication of its text in 1956, mention was frequently made of its (supposed) reading of the article with in John 7:52.1*> A conjecture in this regard had already (without MS support) found its way into the Nestle apparatus. Now, with P66, one had early MS support for this conjecture. J. N. Birdsall in fact, who concluded as to the generally second ary nature of many of its readings and especially its singular readings, allowed as one of the two
i

singular readings in the papyrus which had the strongest prima facie claim to originality.*7 This support by P66 of the previous conjecture,

*5see especially many of his conclusions in "An Eclectic Study. Cf. the list of such readings noted by Metzger in The Text of the New Testament, p. 178. ^See e.g., C. K. Barrett, "Papyrus Bodmer II: A preliminary report," ExpT, LXYIII (1956/57)$ 176; Michaels, "Some Notable Readings,1, . 152*. 17?fre Bodmer Papyrus, p. 17.

231 however, has been shown to rest on doubtful ground.1 article here is the reading of 66*; and it has been pointed out that the singular readings of P66* are all of dubious quality and most likely are the product of the scribe himself, not his exemplar (supra, p. 169). The reading i > 6* therefore has as little textual i value as the nineteenth century conjectures; and even though this reading is contextually to be preferred, and perhaps even what the author intended in terms of meaning, there can be little question that he in fact wrote without the article.^ Such an argument of course does not necessarily condemn all singular readings in all MSS. Colwell and Tune The

have tended to move in this direction, suggesting that all singular readings "are to be ignored in the subsequent stage of manuscript study" 20 Por them the chief value of

the singular reading is "in the initial appraisal of the work of the scribe in a particular MS." 21 To the arguments that singular readings may be

^Pee, "Corrections of Papyrus Bodmer II and The Nestle Greek Testament," p. 68. ^Even if a good early MS were found which had the article, it must continue to be rejected as secondary on the basis of ardua lectio potior.
20n g

Variant Readings," p. 260.

llbid. Colwell has fruitfully pursued this point in his study on "Scribal Habits in the Early Papyri."

original, or that they should be cited in the apparatus crltlcus in case future discoveries nay include a KS which shares the singular reading, Colwell and Tune respond: sufficient answer lies in the high probability that in a tradition as richly evidenced as that of the NT the origi nal has survived in spine group or type of text" (ibid,). From our analysis of P75 in Chapter V, one nay well hesitate at these points. The fact that P75 (sometimes with P45) has eliminated what were once singular or nearly singular readings of at some points where they seem to preserve the original text, and the fact that both have been Judged as careful preservations of a very early type of text, should cause one to allow the possibility that either of them in a singular reading best represents this texttype. It does not necessarily follow of course that the "best representative of the texttype" also preserves the original text, but it does mean that, in answer to Colwell and Tune, the original reading may have survived only in the "best" representative of a type of text, not necessarily in the entire group of MSS. However, all of this to say that if either P75 or . does-preserve the original text in a singular reading, consideration given to such a reading (besides its best answering the questions of internal criticism) rests chiefly on the Judgment as to the generally excellent quality of these two MSS. And it must be granted to "A

233
Colwell and Tune that singular reading by t h e i r very nature are suspect, which Beans that there must be dee i s ire internal evidence In favor of such a reading before I t Is considered as o r i g i n a l . In any case, choosing singular readings on the criterion of an author*a s t y l e , without proper evaluation of the KSS containing such a reading seems to r e f l e c t a faulty method. Author's Style and Original Text. Another d i f f i -

culty with a method vhlch stay lead one t o choose a poorly attested reading, based on an author's s t y l e alone. Is that I t seems t o overlook the very Important p o s s i b i l i t i e s noted by Metzger: "(a) An author nay on occasion vary h i s usage,

and (b) a scribe who was aware of the author's prevailing usage say have altered a reading in order t o bring I t into harmony with t h i s usage." 2 2 To i l l u s t r a t e the problem here, I note two I t e s s from Kilpatrick's a r t i c l e , "Atticism and the Text of the Greek New Testament," which have a l s o been examined In Chapter IV of t h i s study. 1 In the discussion of the tendencies in P66, i t

was noted that in three instances P66 reads the future middle of Joe against the future active of the Neutral tradition (supra, p . l 4 l ) . I t was a l s o concluded that t h i s

The Text of the New Testament, p . 178.

23* 1 a secondary tendency in P66. In contrast to this, Kilpatrick (pp. 132-133) has argued that Inom is an Attic form, Jijoopat non-Attic. Thia being the case, "we would expect the New Testament writers to use ifaopat." He then illustrates his argument froa the Gospel of John by setting out, with the KS e ride nee, the alx instances where the future of lam occurs. The heart of the Neutral tradition (P75 L) reads the active four tlees (5:25; 6:57, 58; 14:19). In three of

these instances they are variously Joined by D and embers of the Byzantine tradition. On the basis of the witnesa of P45, P66, and P75 (the former two read the active once each), Kilpatrick notes that the active occurs in KSS which are older than A.D. 200. He therefore concludes: As the variation came into being in the second cen tury, the century of Atticisa, it is more probable that the evangelist at the end of the first century used the non-Attic middle which was later corrected to the Attic active future. That the evangelist should go out of his way to introduce an Attic fona into his Koii^e Creek which the second century scribes then changed to the Koine form seems most unlikely. Ve may accordingly regard the middle-future as what the evangelist wrote and the active as an Atticist correction of the second century. (P. 132) Kllpatrlck further notes four places in the Epistles where the future active occurs, in three of which he finds Atticism at work (Rom. 6:2 - ; II Cor. 13:4 Jtioopcv - ; James 4:5 Snooper - ). Heb. 12:9 is almost universally attested. Only at

235
Of these last four readings, however, two things should be noted: (l) They all occur in the first person

plural, and these are the only occurrences of the first plural of the future of lam in the NT. (2) Only one of

these readings reflects a variation between the middle and passive. It would seem, therefore, that the future active
s

in the first person plural is the fixed form, and that it was the future middle which was being resisted by the MS tradition, not vice versa. But the real weakness in Kil pa trick's argument is that he fails to note the other ten occurrences of the future of lam in the KT (Ht. 4:4; 9:l8; Luke 4:4; 10:28; Rom. 1:17; 8:13; 10:5; Gal. 3:11, 12; Heb. 10:38), all of which occur, as in John, in the second or third person. In every Instance but one, Luke 10:28 where D and 28 read the active, occurs in the future middle. The fact that in John the future middle is univer sally attested (except for P45) at 11:25, and the active attested by the majority of uncials at 6:58, plus the universal attestation of the future middle outside of John, seems to point to a conclusion directly opposite that of Kilpatrick. The question which Kilpatrick must answer is, why only in John do the Alexandrians, and others, reflect Atticist tendencies? For whether or not the evangelist "went out of his way" to do so, the future active of a is strictly a Johannine phenomenon in the NT. Furthermore,

236
the combined evidence of 6:58 and 11:25 indicates that the Fourth Evangelist was not consistent. What we have then in John is not P66 and later MSS resisting "corrections" toward Atticism, but P66 and later scribes resisting the "Atticism" of the Fourth Evangelist in favor of the more common biblical idiom.
*

The tendency of "careful preservation" in P75, as well as the "mixed" nature of the text of P66, especially after John 5 seems to lend weight to this conclusion. 2 Kilpatrick has also argued that the ano%px9tx^

etnev idiom, where there are variations, is to be preferred to either of the verbs occurring by itself; for "no Greek of any period, left to himself, would say or write " (p. 126). He therefore concludes: Hence we are not surprised when we find that often where and the like occur in our Greek text there are variants designed to mitigate or remove this unGreek expression. We may even suspect that sometimes the attempt to improve the language has been successful and that the more Greek expression is in our text and the original unGreek wording in our apparatus. (Ibid.) However, our discussion of this idiom in John (supra, pp. 148 ff.) seems again to point to the opposite conclusion, namely, that the author himself had a basic unGreek idiom, [] xat [awco] which he himself altered occasionally toward the "more Greek" xptGq []; the scribes, on the other hand, tended not "to mitigate" the author's unGreek idiom, but tended rather

237
to conform to that idiom where the author himself had used the "more Greek" form. She following evidence seems to substantiate such a conclusion: (a) The full idiom occurs without B S variation 19 I times in John. 23 on nine other occasions a single MS

"mitigates" this unGreek form. (b) The "more Greek" idiom (less xat etnev) occurs 21 times without MS variation . 2 5 If one may trust the MS evidence at all, these two sets of readings clearly indicate that the author of the Gospel himself used both the "Attic" and the "non-Attic" forms of the idiom. Moreover, the nine instances where a

single MS reads the "more Greek" for the "unGreek" idiom indicate that there is very little tendency in the MS tradition to mitigate the unGreek idiom. (c) On the other hand, on 17 other occasions, where the majority of MSS read without xat etnev, one or a few

1:48, 50; 2:18, 19; 3:9, 27; 4:10, 13; 7:l6, 21, 52; 8:38, 48; 9:30, 34; 14:23; 18:25, 30; 20:28. *3:3 (K* om. xat etnev): 3:10 (083 om. ) 4:17 (* om. xat etnev); 5:19 (* . . . ) 6:43 (053 om. xat etnev); 6:26 (N om. xat etnev 8:14 (& etnev ); 12:30 ( om. xat etnev); 13:7 (33 om. xat etnev). 5:7; 6:7, 68; 7:47; 8:34; 9:20, 27; 10:25, 33; 11:9; 12:34; 16:31; 18:8, 20, 23, 34, 35, 36; 19:7, 15; 21:5.
25 2I

23

238 MSS read the full idiom. The MS evidence for those which read the full idiom at these points is given below: 3:5 6:70 7:46 8:19 L 053 69 124 174 213 230 579 1093 124l pc (D) 1187 a ff2 sa 544 c bo D 27 78 543 700 713 1093 1188 1241

8:33 D 1 7 291*565 658 660 1293 1354 1582 2193 j ! 8:49 8:54 9:3 291 440 565 1093 1170 1093 1170 1242 1555 bo 053 565 e b

10:32 33 10:34 66 D 13:8 1071 r 1 aeth syPal

13:26 D 13 346 13:38 D aur c ff2 18:5 X 213 f

18:37 P66 19:11 a c ff2 syPal ! 19:22 1170 1242 (d) On only four occasions do the Neutral MSS (always with several others) read the shorter form against the majority (l:49; 7:20; 9:11, 25). Prom these sets of figures the conclusion seems unmistakable that the MSS tend not to become "Atticist" against the author, but rather that they tend to conform to the more characteristically biblical idiom. The fact that

239
the KSS which do have variations at this idiom are not necessarily "superior" witnesses seems to confirm this conclusion Codez Sinaiticus, for example, whose tenden-

cies to be "wild" in John make it suspect, "mitigates the unGreek idiom" five times, but it also reads the full idiom against most of the r^st at four other places. The fact that P66 (almost alone) conforms to the more characteristically Johannine idiom at two places indicates that even in the second century (Kilpatrick's "Atticist" century), the tendency to "conform" is at least as great as any tendency to "Atticize," as far as this idiom is concerned. The Evaluation of Variants Where Internal Questions Are Indecisive. The problem here has already been noted in The inadequacy of "rigorous

Chapter IV (pp. 121-122).

eclecticism" at this point may best be illustrated by referring again to Grant*s three dicta for evaluating variants: 1. No one type of text is infallible, or to be preferred by virtue of its generally superior authority. 2. Each reading must be examined on its merits, and preference must be given to those readings which are demonstrably in the style of the author under consideration . 3 Readings which explain other variants, but are not contrariwise themselves to be explained by the others, merit our preference, (supra, p. 228) The problem for "rigorous eclecticism" here is, on what grounds does one choose when rules 2 and 3 point to directly opposite conclusions? It would seem that when

such happensand it often doesthat rule 1 must be laid

240 side; for "the generally superior authority" of a given KS or type of text ay be the deciding factor. An illustration of this is the set of variations in John 5:17* noted in Chapter m . Without regard for the

moment to the external evidence, the following variants are found:

/

1. 2. O 6c 3. 6e h 6c

arrcxot6il anexettq cireptraro atttxpi^ Beading 1 i s preferred

The readings are here l i s t e d in the order of preference according t o Grant's rule 2 . style. because i t i s the only reading i n accordance with Johaxmine The use of 6c for the continuation of narrative i s so uncommon in the Gospel of John as t o make the "rigor ous e c l e c t i c " suspect I t a t any point; and 6c a t such a point Is found in the KT only here and in a reading of 33 579 PC In Mark 15:13On the other hand, according t o Grant's rule 3 the exact opposite i s the order of preference. The 6c i s The the only reading which explains the other variants, and i s not contrariwise i t s e l f t o be explained by the others. ore common form of expression; the omission of either r e f l e c t s a preference for a more c h a r a c t e r i s t i c a l l y Johannlne mode of expression. IT the omission were original, one can scarcely explain the addition of e i t h e r 6c or 6e i s e a s i l y explained f r o as the preference for a

241

6cespecially with a font of in the Fourth Gospel. I f 6c were original, one can explain the omis

sion as a conformation to Johannlne s t y l e , but the 6c defies explanation except as an Inadvertent scribal error. I t would seem, therefore, e i t h e r that Grant's rules must be transcended, or e l s e a choice must be Bade between rules 2 and 3However, when external evidence la added, a decision In favor of 6c seems well founded. dence Is as follows: 1. 2. ancxp8ti D E F E K S U Y T pier TR 6c t 66 C L K K G A A O p a 6c * W 6c anexptflti 75 The reading of A i s unexpected The external e v i

I.

The textual relationship between P75 and means that theirs i s a single witness.

support, inasmuch as i t more often reads with the Byzantine KSS than with P75 B, where these two types of t e x t d i f f e r . Moreover, our Judgment of P75 and i n Chapter V i s that theirs i s a witness t o the faithful preservation of a given type of t e x t . On the other hand, P66 has been noted f r e

quently to d i f f e r froa P75 in favor of an easier reading, and thus i t does here. Furthermore, D, the only early

witness to the more "characteristically Johannine" reading, has been noted often t o conform to a "later" t e x t (supra, pp. 49-50). When, therefore, the e a r l i e s t and "best" KSS have

242 the reading which beet explains the others, this combina tion would seem to overrule the choice dictated by "rigor ous eclecticism." There are many other examples where the combination of early and "best* authorities with the canon ardua lectio potior seems to overrule the principles of rigorous eclecticism. The following are but a sampling:

1. John 11:12. One of John's introductory formulae has the following variations: a. b. e. d. . f. etnav etTTav etnav etnav tnav etnav ot avmc ot ot ot D 579 b ff2 251 254 a c 266 P75 C pc 44' 122 7 116 314 1200 1 L Byz aur e f TR

On the basis of our examination in Chapter IV (supra, p. 155), the most characteristically Johannlne reading here is (a). The combination of verb-object-subject (VOS) is far more common than the combination VS in (e) and VO in (d). But the only reading which explains all the others and Is itself explained by none is (c). This is easily the ardua lectio inasmuch as in the some 120 occurrences of this idiom in John the order VSO occurs without variation only where the object is a noun (e.g., 2:5; 6:67; 18:11) or where the prepositional phraee . . . replaces the dative (e.g., 4:48; 8:57; 11:21). The order VSO where the

pronoun is in the dative occurs only here and at 11:44

243
(P75 C L V aa bo e vg) in the Gospel of John.2** Therefore, reading (b) may be dismissed as a modifi cation of (a), and readings (a), (d), (e), and (f) all represent various attempts to remove the difficulty of the word order of P66 P75 et al. The fact that this reading is supported both by the earliest . r d "best" MSS, as well ai as by various others which do not necessarily have "Neu tral" proclivities, seems to indicate that the "more difficult" and "best supported" reading is to be preferred. 2. John 12:16. The MS tradition has the following

variations: a. b. c. ot ot 75 579 66 A D W L Q rell TR 265 1219 1346

On the basis of Johannlne usage, reading (b) is to be preferred. Of the 33 occurrences of with the

possessive in the Nestle-Aland text, only here do arty of the KSS have the possessive first. Moreover, of 15 other

occurrences of in Nestle-Aland, where various MSS add the possessive, they always add it after the noun. However, the canon ardua lectio potior favors the reading of the possessive first. There seems to be no

other logical explanation of this reading except as sheer error. However, that the "error" is found in secondary
2

"Thie unusual order also occurs in three singular readings (6:32 579; 7:33 1241; 9:41 D ) .

244 witnesses to the Neutral tradition ( 579) and as well as In the "purer" witnesses to the Neutral tradition Indicates that it was both an early and somewhat widespread "error." Here again MS evidence supports the "more difficult" read ing as against the "more typically Johannine." Besides in such readings as these, where an author's style is Involved, external evidence seems to be particu larly crucial in those instances where neither of Grant's rules 2 or 3 are applicable. A case in point is the prob

lem of the mission of the seventy or seventy-two in Luke 10:1 (10:17), a problem which Metzger has already discussed thoroughly. ' Lucan style is simply not a consideration here. Moreover, the reasons for choosing either eeventy

or seventy-two in terms of the background of these two numbers seems evenly balanced. At least neither reading is

"more difficult" than the other, and, as Metzger has shown, either may well explain the existence of the other. The present writer happens to prefer seventy-two, on the grounds that there seems to be more background (at least biblical background) for the number seventy. This

would make seventy the more common number, and therefore the one to which a scribe is more likely to have changed. Moreover, the number seventy-two seems to require no more

*'"Seventy or Seventy-two Disciples?" NTS, V (1958/59), 299-306.

245
extensive background than the immediate context of the GospelsIt is simply a multiplication of "the twelve." Metzger has suggested that this "may be the result of a mathematical penchant of an Alexandrian recensionist who altered '70 to 72 for, so to speak, scholarly reasons" (p. 305) But this begs the important question as to whether there was in fact such a recensionist in Alexandria at such an early date as ca_. 150-175. And why could not Luke have had such a "penchant" as well as a second century scribe? The choice thus far however is almost totally a subjective one. When one considers that seventy-two has the support of P75 D R 40 1604 It vg s y 8 0 , 2 8 such early and widely distributed evidence seems to point back to the Lucan original rather than a "scholarly" Alexandrian recensor. We may conclude, therefore, that "rigorous eclecti cism" Is both a limited and sometimes short-sighted method. It is limited in that it is not capable of arriving at a decision In cases where the questions of internal evidence are opposed or evenly balanced. It is short-sighted in

that it falls to give proper weight to the historical ot f co Most critical editions also include the witness of 45, on the basis of Kenyon's edition. Metzger has questioned the reading of P45 here, so it has not been Included.

246
evidence supplied by the MSS themselves. 9 Our analysis of P66 and P75 has indicated that MSS ay be evaluated as to their general "worth" in terms of their witness to the original NT text. Moreover, such an

evaluation is often helpful in making a decision on Individual readings and must therefore be a part of one's total method. The problem here is that one must free himself

from letting his evaluation of the MSS prejudge the final decision. It is at this point that contemporary eclecti-

cism of the less thorough type wishes most to part with Eort. To the degree that eclecticism is able to decide variants without a prejudgment on the basis of external evidence, it would appear to be a valid methodology. There remains, however, the problem of the "Hortian face" of the results of our non-Hortian method. II. P66 AND P75 AND EARLY TEXTUAL RECENSION

Apart from their general importance as external evidence, perhaps the most significant contribution of the

the evaluation of this form of eclecticism by K. Aland in "The Significance of the Papyri for Progress in New Testament Research," The Bible in Modern Scholarship, ed. J. Philip Hyatt (Nashville, 1955), p. 340: "Seeing those eclectics at work who seem to regard the language of the NT writings as fixed and who merely seek in the tradition of the text the groups of manuscripts, or the particular manuscript, or the version, which they need for their text, I must doubt, not only the scientific value of their method, but also the principle upon which that method is based."

247 cooblned witness of P66 and P75 is to the question of early textual recension, and especially to the question of the "scholarly Alexandrian" nature of the alleged recension of the Neutral texttype. The Keutral Texttype as Recension. The steps lead ing to the current, almost universal acceptance of the recenalonal nature of the Neutral texttype have already been briefly sketched (supra, p. 9, n. 17). Our interest here is to focus particularly on two cajor causes in this century for the moveaent away from Hart toward Bousset, namely the papyrus discoveries up to I960, and the argu ments of Kenyon and Zuntz. Probably the leading factor in current disavowal of Hort's "pure line of descent" froa the "original" NT to was the papyrus discoveries of this century, up to and including the discovery of P66. All of these discoveries shoved a such more fluid and "nixed'' state of textual transmission than Hort had proposed. In fact the mixture

was of such nature that none of the fourth century texttypes was found in these KSS in a "pure'* state. This led to such expressions as "pre-recenslocal* and "protoAlexandrian." Typical of this "new" attitude elicited from the papyrus discoveries were Kenyon*s remarks in the Introduc tion to his edition of the Chester Beatty Papyri:

2*8
This much . . . ay be said without hesitation. On the one hand, it [the Chester Beatty discovery] is not an out-and-out supporter of the Neutral* or Vatican type of text; but neither is it, on the other hand, an outand-out supporter of the 'Western' type. . . . ?or the moment it crust suffice to point out that the occurrence of this type of text in a manuscript from Egypt con temporaneous with, or at latest not arach later than, Origen . . . points, perhaps decisively, to the con clusion that the Vatican MS does not represent a text of original parity? dominant in Egypt throughout the second and third centuries; . . . and that the Vatican text represents the result, not of continuous unaltered tradition, but of skilled scholarship working on the best available authorities.30 Along with the papyrus discoveries, plausible hypotheses of the recensional process In Alexandria were also forthcoming. In his article in the Iagrange Fest

schrift, Kenyon proposed: -During the second and third centuries, a great variety of readings came into existence throughout the Chris tian world. In soee quarters, considerable license was shown in dealing with the sacred text; in others, more respect was shown to the tradition. In Egypt this variety of texts existed, as elsewhere; but Egypt (and especially Alexandria) was a country with a strong tradition of scholarship and with a knowledge of textual criticism. Here, therefore, a relatively faithful tradition was preserved. About the beginning of the fourth century, a scholar nay well have set himself to compare the best accessible representatives of this tradition, and so haye produced a text of which is an early descendant.-*1

Rochester Beatty Papyri, General Introduction, p. 16. Henry A. Sanders (Tne Egyptian Text of the Four Gospels and Acts," HTO, XXVI [1933J, 77-98) argued from the evidence of the papyrus discoveries that there was an almost com plete supremacy of the Western text in the third century. Although his argument left many unconvinced, the nature of the data certainly made possible such an argument. ^"Hesychiua ana the New Testament," p. 250.

249
JCenyon suggested further that the homogeneous character of throughout the MT also pointed to "scholarly recension." Since the books of the MT originally traveled

as individual units and then probably in various corpi, these units and corpi quite naturally have different tex tual histories. In the Gospels, for example, Mark has the largest a r ~ r t of variation per page of text, while John a . has the least. Moreover, the Western text exhibited in D

has much greater variation in Luke-Acts than it does in Kark or John. Kenyon therefore concluded that the charac ter of B, which is so homogeneous throughout the NT, implies "the exercise of editorial selection" across the various textual histories, as well as the internal editing of the various texts.*2 Kenyon*s conclusions about became a byword in NT textual criticism. The recent text critical handbooks33

and NT Introductions,^ as well as articles on "trends" in

^The Text of the Greek Bible, p. 205. 33e .g., Vincent Taylor, The Text of the New Testament (London, 1961), p. 54; Metzger, The Text of the New Testament, pp. 215-216; Greenlee, Introduction, pp. bo-b7. ^ e.g., A. H. McNeile, An Introduction to the Study of the New Testament, 2nd ed. rev. C. S. C. Williams (Oxford, 1953), p. 434; Felne-Behm-Kummel, Introduction,

p. 384.

250 text criticism35 and on texttypes,36 were almost unanimous In their concurrence with Kenyon's conclusion that the Keutral text "is now generally regarded as a text produced

in Egypt and probably at Alexandria under editorial care."37 Whereas such a recension was usually thought to be the work of one hand, more recently it has been viewed as the result of a long process. Gunther Zuntz's reconstruc

tion of this process shows erudition as well as sensible imagination, and is worth noting in full. Its beginnings were inconspicuous, and roughly 150 years passed before it culminated in the 'Euthalian' edition. Prior to this final achievement, the Alexan drian correctors strove, in ever repeated efforts, to keep the text current in their sphere free from the many faults that had infected it in the previous period and which tended to crop up again even after they had been obelized. Theee labours must time and again have been checked by persecutions and the confiscation of Christian books, and counteracted by the continuing currency of manuscripts of the older type. None the less they resulted in the emergence of a type of text (as distinct from a definite edition) which served as a norm for the correctors in provincial Egyptian scrip toria. The final result was the survival of a text far superior to that of the second century, even though the revisers, being fallible humans, rejected some of its correct readings and introduced some faults of their own.38 Zuntz, however, was candid to observe that "even so,

'e.g. K. . Clark, "The Effect of Recent Textual Criticism,1 p. 37i K. Aland, "The Present Position," p. 750. 36colwell, "The Origin of Texttypes," p. 137.

37The Text of the Greek Bible, p. 208. 38rhe Text of the Epistles, pp. 271-272.

251 It must be admitted that no direct evidence attests the philological endeavors which we have Inferred" (p. 272).

He finally rests his argument on the acknowledged philological skill of Orlgen and on Clement's apparently Intimate acquaintance with grammatical terminology. He concludes

that "the Greek grammatical tradition and technique was among the pagan achievements by which Alexandria enriched the Christian tradition" (p. 273). Perhaps the most significant part of Zuntz's hypothesis is his anewer to the question as to how the Alexandrian copyists achieved the high quality of their texts. Against those who had suggested that It was the result of expert collation and editing of "wilder" texts, Zuntz cogently observed that such a process could only result in the "emergence of an average text of that very type" (p. 273). Rather, he maintained: "This Indeed is the essential fact:

somehow the Alexandrian collators must have been enabled to use manuscripts superior to those current in the second century" (p. 274). The preservation of such texts as these, Zuntz further noted, "bespeaks the conscious appreciation of the original wording as a value per ae. It has already

been pointed out that this is not the attitude of the believer or the theologian as such: it requires at least a

touch of the philological mind" (p. 275). Such, then, was the status of this question and the most significant hypothesis to answer it up to the discovery

52

of P66 and P75

Our present concern is to indicate the

significance of these MSS to this question of the receneional nature of the Neutral texttype. But before so doing, a word is in order about the term "recension" itself. Recension and Texttype. In his Ou en est la critique

textuelle du Nouveau Testament Jean Duplacy has noted that the term "recension" is often used with imprecision.39 the one hand, it has been defined as "the text created,
i
!

On

according to specific principles, for use in one segment of

t

the church."**0 term.

This is the more strict definition of the

On the other hand it is also used for "editorial

activity" of any kind where a MS shows conscious revision or correction. As long as texttype are considered as the work of an individual such as Lucian, Hesychius, or Origen, the first definition means the actual work of that individual in "creating" a text. In recent years, however, texttypes When |

have been considered to be processes of transmission.

texttypes are thus considered, "recension" may have refer ence to the beginning of that process, its culmination, or
i

^Paris, 1959, p. 33. *This is the definition used by Porter, "Papyrus Bodmer XV," p. 364, n. 7. Jt appears to be a translation of Duplacys translation (Ou en est, p. 33) of Paolo Sacchi, Alle origin! del Nuove Testamento: Sagglo per la storla della tradlzione e la critica del testo (Firenze, 195t>;, p. 35.

253
any kind of editorial activity along the way. this point that more precision is needed. If in fact it ie correct to speak of texttypes as processes, then the term "recension" in the strict sense of a "created text" seems to be incorrect, unless it can be demonstrated by reasonable hypothesis that the "process" had a definite beginning.*1 Otherwise it would seem to be It is at

more precise to consider our texttypes as "textual traditions" with "recensional activity," rather than as "recensions." In the lese strict sense the adjective "recensional" may be retained to refer to a MS such as P66, which shows clear signs of "editing." Whether the scribe of P66 waa

intending to create a definitive text for one segment of the church may never be known. What is_ knownpresently at leastis that his text was not definitive enough to be followed closely by any other known MS. But whether definitive or not, his text is "recensional" insofar as he deliberately altered it when making corrections. P75 and the Neutral Texttype as Recension. The analysis in Chapter V of P75 in Luke, together with Porter's

^ C f . the work of Hetzger, "The Lucianic Recension of the Greek Bible," Chapters in the History of New Testament Textual Criticism (Grand Rapids, 1903) PP. 1-41. Metzger here argues that Lucian does in fact stand at the beginning of the Byzantine texttype. If so, then his work is recensional in the strict sense of the term.

25* analysis of it text In John, has given proof beyond any reasonable doubt that If the Ueutral texttype Is a recen sion in the strict sense of that term, it is not a recen sion created in the late third century. P75 has forever laid to rest the "Besychian hypothesis." Furthermore, the close relationship which exists between P75 and demon strates that the text of is not the late culmination of a process. Zuntx, therefore, seees to be partially in error when he epeaJcs of the Neutral texttype as a long process with inconspicuous beginnings, which after roughly 150 years "culminated in the 'Butnallan' edition;" for that "Buthalian" edition already existed in the second century in the text of F75. The only "process" involved between

the time of P75 and is that of carefully preserving a given text. Such activity as this is not "recensional" in any sense of that tern. Moreover, it is important to note from the analysis In Chapter V that the same essential relationship between P75 and exists in both Luke and John. This means that the Neutral texttype existed in the second century across two distinct textual histories. Therefore, if the homoge neous nature of this texttype is due to good "selection" as well as good "recension," as Kenyon argued, then such
up

*The assumption here of course, as with Zuntz, is that best represented that "Euthalian* edition in the fourth century. See supra, p. 121, n. 3.

255
selection also bad to bare taken place in the second century. Finally, and quite significantly, It aay be shown by the scribal quality of P75 that this KS Itself was not that recension. Although the nuaber of scribal errors In the text Is ninipal, those which do exist are of such nature that one can be certain that we hare here a copyist, not a recensor. Both the large nuaber of uncorrected errors (approximately 76) and the fact that the great sajority of the corrections are of obvious scribal errors point to such a conclusion. As Colwell has pointed out: In P75 the text that is produced can be explained in all Its variants as the result of a single force, namely the disciplined scribe who writes jrith the Intention of being careful and accurate.^ It say be concluded quite positively, therefore, that the so-called "Euthalian" edition of the Heutral texttype was already In existence in Egypt in tfce second cen tury. This does not asean that it was necessarily the "doainant" text of Egyptindeed, the other papyri suggest quite the oppositebut it does nean that the texttype per se existed; and if the relationship of L C 33 etc. can be called a texttype at all, then P75 is the earliest known nember of this texttypeand it is a full-fledged

^"Scribal Habits," p. 381.

256 member.
The question which must be asked, then, is whether the Neutral texttype Is recensional at all. If so, there

emphasis here is directed toward what appears to be a curious conclusion on the part of Aland in his article, "The Significance of the Papyri," pp. 336-337. He acknowledges that "P75 shows such a close affinity with Codex Vaticanus that the supposition of a recension of the text at Alexandria, in the fourth century, can no longer be held." But then he goes on to suggest that the fourth cen tury Alexandrians did "apparently . . . take those manu scripts which they regarded as reliablefor Luke and John a manuscript such as P75and to correct, text errors and corruptions or what they thought to be Such." But this seems to fall to acknowledge how close the relationship be tween P75 and actually is; for it is at least as close as that which exists between the second level of relationships in Family 1. It seems curious to this writer that one may speak easily of a B text, as if this represented a textual onolith, yet feel that some sort of "revision" transpired between the text of P75 and B, whose relationship is so close that is a distant cousin in comparison. This whole question seems to be obscured further by Aland when he next suggests that: "Of course, one can speak of an Egyptian or an Alexandrian text-form, as well as of an Antiochian or Byzantine text-form. . . . These are, it seems to oe, the only text-types which may be re garded as certain, and that only since the fourth century. . . . It is impossible to fit the papyri, from the time prior to the fourth century, into these rwo text-type3, to say nothing of trying to fit them into other types, as fre quently happens. The simple fact that all these papyri, with their various distinctive characteristics, did exist side by side, in the same ecclesiastical province, that is, In Egypt, where they were found, is the best argument against the existence of any text-types. . . . " Italics mine. This seems not to take the witness of P75 seriously. I am inclined to agree with Colwell, when he suggests that "in dating texttypes, what is needed is a datable witness to the type, not only of some of its readings" ("The Origin of Texttypes," p. 138). P75 Is such a datable wit ness, and not merely to some readings, but practically to the whole structure. If the term texttype is to mean any thing at all, then P75 i a*member of a texttypeand is so In the second century.

257
seem to be only two alternatives. Either It was a recension created In the second century, or else it was the culmination of a process, but a process which had very little time to develop. In either case the text, as Zuntz

has argued, almost certainly had to be "made" in a "philological ly conscious" center such as Alexandria. The final question to be asked in this study is whether historical probability favors either of these two alternatives, or whether it favors a third alternative that the Neutral texttype is a carefully preserved tradition and not a recension at all. Origen and the Neutral Texttype as Recension. As

noted above, Zuntz argued that "the very existence, amid the welter of 'wild texts, of manuscripts of a pure type bespeaks the conscious appreciation of the original wording as a value per se." Such an attitude, he reflected, was

not necessarily required of a believer as such, but it did require "at least a touch of the philological mind."^5 existence of such a philological mind in the church of Alexandria has been posited chiefly on the well-known mastery of the philological tradition on the part of Origen. Two things, however, should be noted in this regard. In the first place, P75 and Origen have, in the Gospel of The

*5zuntz, The Text of the Epistles, p. 275.

258 John, closely related texts.^> m i s means that, since P75

antedates Origen, he is merely witnessing to this text; he is not responsible for it. The philological tradition, therefore, capable of creating such a textif Indeed there was such a "creation"had to have been a part of the church of Alexandria* before Origen.^7 Secondly, Metzger has recently analyzed the "Explicit References in the Works of Origen to Variant Readings in New Testament Manuscripts."^ m s conclusions are important to the argument of this study: He was an acute observer of textual phenomena but was quite uncritical in his evaluation of their significance. In the majority of cases he was content merely to make the observation that certain other copies present a different reading, without indicating his preference for one or the other variant. This tantalizing nonchalance is so unlike his careful procedure in dealing with the Greek text of the Old Testament that some special explanation must be sought.. . .

This is based on a personal collation of P75 and the text of John in Origen's Commentary of John (A. E. Brooke, ed., The Commentary of Origen on S. John's Gospel, 2 vols. [Cambridge, lb9b}). See also R. V. G. Tasker, "The Text of the Fourth Gospel Used by Origen in His Commentary on John," JTS, XXXVII (1936), 146-155; and K. W. Kim, "Origen's Text of John in his On Prayer, Commentary on Matthew, and Against Celsus," JTS, n.s. I (1950J, 74-54. ^'Zuntz in fact has so concluded: "The conclusion is almost inescapable that already in the latter half of the second century the Alexandrian bishopric possessed a scriptorium, which by its output set the standard for the Alexandrian type of Biblical manuscripts" (The Text of the Epistles, p. 273). ^ I n Biblical and Patristic Studies, ed. J. N. Birdsall and R. W. Thomson (Freiburg, 19&3), PP. 78-95.

259
t

On the whole his treatment of variant readings is most unsatisfactory from the standpoint of modern textual criticism. He combines a remarkable indifference to what are now regarded as important aspects of textual criticism with a quite uncritical method of dealing with them. (Pp. 93-94) We may conclude two things therefore about Orlgen:

(l) In the Gospel of John he is a witness to the Neutral texttype; but as such he represents the "preservation" of euch a type of text, not its "recension." (2) In contrast

to his work on the Old Testament, Origen never shows a concern for a "critical text" of the New Testament writings. If such an acknowledged "textual expert" as Origen showed no particular interest in "scholarly recensional activity" in the early third century, one may well ask whether historical probability favors the existence either of the person or the incentive to create a text on the principles of scholarly recension in the second century. The evidence seems rather to point to "careful preservation," probably with occasional stylistic changes, as the true product of the Alexandrian philological expertise. P66 and the Neutral Texttype as Recension. The

tentative conclusion as to the non-recensional character of the Neutral texttype based on the evidence of P75 and Origen is further substantiated by the evidence from our analysis of P66. The primary importance of the MS at this point is in the corrections to its text; for herein is our earliest piece of actual historical evidence in which

260 recenslonal activity is clearly present. Our analysis of the corrections of ?66 led to this unmistakable conclusion: When one does find recenslonal activity in Egypt at the end of the second century, such activity is not guided by apparent controls, it is not the work of "scholarly" recension. Furthermore, the nature of the corrections does not point to concern over the preservation of the original text. Here, quite in contrast to P75, but in keeping with the general attitude of Orlgen, one has no apparent concern for the original wording per se, but for the best "sense" of the text. This evidence from P66 of course does not mean that recenslonal activity of another kind did not exist. But it does give direct evidence of one kind that did in fact existand at a very early date and probably in Alexandria.*9 The contribution of "Alexandrian philological knowhow," therefore, appears to have found its best expression in the careful preservation of a given text (P75 and B ) . Such preservation is probably guided by the "conscious

is conjecture, of course. It is based on the fact that the calligraphy of the K3 is of such high quality and that the corrections were made from a second MS, which may indicate the work of a scriptorli. Cf. Colwell, "Scribal Habits," p. 382: "P66 gives the impression of being the product of a scriptorium, i.e. a publishing house. It shows the supervision of a foreman, or of a scribe turned proofreader."

261 appreciation of the original wording as a value per so." But when one finds actual reeenslonal activity, even in Egypt at an early date, it does not differ earkedly from the kind of reeenslonal activity one finds at a later date in the process of textual transmission called the Byzantine texttype.
s

"

The Neutral Texttype and the "Original Text". The combined evidence of the "carefully preserved text" of P75* the "non-scholarly recension" of P66, and the lack of "editorial" concern on the part of Orlgen seeas to point to one conclusion: The Neutral texttype does not represent * recension but rather one for of preservation of the original text of the NT. This does not Bean, however, that it is a necessarily "pure" form of preservation. The quality of preservation which is found in this texttype nay only he demonstrated by a careful, direct comparison, based on the principles of reasoned eclecticism, of the KSS of this tradition with other KSS and textual groups at all points of variation over a large section of text. A study of this kind, similar to that of Zuntr for ?kS and of Sakae Kubo for P72 and B,^ appears to be the next eajor task called for in regard to P75 and B. Although such a study is beyond the compass

1965).

^ g 7 2 and the Codex Vatlcanus (Salt Lake City, (Studies and Documents, ho. 27)

262 of this present work, some "consensus* suggestions as to the generally excellent quality, and therefore "relatively pure" form of preservation, of the primary witnesses of the Heutral texttype are in order. 1. The analysis of the text of F66 in Chapter IV, limited though it was# in the number? of variants analyzed, indicated that in the great majority of instances the text of P75 reflected a superior text; and this superior text was usually considered most likely to represent the Johannine original. 2. The work of Kubo on P72 and in I Peter,

II Peter, and Jude also indicated that the Keutral texttype preserves a generally superior text. Be concluded, simi larly to our tentative suggestions as to the relationship of P75 and B, that where they differ (apart from singular readings) "P72 as a whole has a text superior to that of B* (p. 152). But, he continued, "in saying this, we must be aware that the comparative quality of the text of as a whole is not affected" (ibid.). His final conclusion was that "where P?2 and agree in their basic text, their common text is almost always superior to any other opposing combinations" (p. 15^) 3 * collation of the major early Greek KSS with some of the critical texts of the past century indicates that the consensus of textual criticism far more often favors the text found in P75 and than that found in any

other single M or textual tradition. S Such a collation of the texts of Westcott-Hort, Borer, von Soden, Nestle-Aland and the NB a t the 70 variation-units in Luke 10 yielded the following s i g n i f i cant items: (a) The highest percentage of agreement with any

ancient M was between Westcott-Hort and (9). S (b) The lowest percentage of agreement of any of

the c r i t i c a l t e x t s with was 74# (von Soden). (c) The highest percentage of agreement between any

of the c r i t i c a l t e x t s and a M outside the Keutral tradi S t i o n was between Bover and D (55#) Similar analyses such as t h i s by K. V. Clark^1 and K. Aland^2 indicated similar r e s u l t s . dominance of the Neutral t e x t . Proa t h e i r r e s u l t s both Clark and Aland concluded that we are s t i l l under the Perhaps s o , but i t ay a l s o ean nothing more than that even in a text chosen verse by verse, under e c l e c t i c principles, such a text w i l l look ore l i k e that of P75 than of other KSS simply because t h i s M tradition has more f a i t h f u l l y preserved a good text S than the others.53
51

"The Effect of Recent Textual Criticism," pp. 30-

37.
^"The Present Position," pp. 718-722. 53consensus, 0 f course, is not a foolproof guide to general excellence, but it does indicate the high esteem

264
I If thee various Judgments are correct as to the cooperatively excellent quality of text found in the Neutral K3S, then we may make the following general conclusion about this texttype: The MSS of the Neutral tradition, rather than reflecting careful editing, reflect a "rela tively pure" form of preservation of a "relatively pure" line of descent from the "original" text of the NT. Although this conclusion sounds suspiciously Hortian, it is so only incidentally. The chief concern here is not

with the validity of the Hortian textual theory, but rather with the validity of the current method of "reasoned eclec ticism" which trie8 to use a balanced judgment based on both external and internal criteria but relying exclusively on neither. To the question of the validity of this method

we now turn by way of conclusion. ; III. CONCLUSIONS: "REASONED" ECLECTICISM AS A VALID TEXT CRITICAL METHOD

In the Introduction to this study, it was noted that the chief cause of unrest in current textual criticism was the sense of need for a new, valid method. A "new" method

was called for because (l) the "old" method had basic

with which this textual tradition has been regarded. Even Kilpatrick, who is more willing than others to lay aside MS evidence In favor of internal evidence, has acknowledged: ". . . the Alexandrian text and especially are our best authorities" ("Western Text and Original Text in the Gospels and Acts," p. 36).

265
limitations which inherently condemned it as a total method and (2) the "current" method, in spite of new discoveries and its rejection of the "old" method, resulted in a text which for all practical purposes was indistinguishable from that of the "old" method. The dilemma of the current

situation was that both textsthat of the "old" and the "current"had dominantly "Neutral" features. For WestcottHort this was no problem; the Neutral text was considered "neutral." For current criticism this was an anomaly; for the Neutral text at best was considered to be a well-edited text of the third century. The thesis of this study is that the "current" method, namely "reasoned" eclectism, is in fact a valid method, and this for the following reasons: 1. The analysis of P66 and P75> and the conclusions in this chapter, indicate that if eclecticism is a valid method it must be of the "less thoroughgoing" or "reasoned" type, rather than of the "rigorous" type found in the work of Kilpatrick. 2. The "Hortian face" of our resultant texta j

major cause of unrestis quite incidental to whether or not the method is valid. A text which more often has

Neutral readings than readings from other MSS and texttypes does not necessarily mean that our method is under the dominance of this text type. It may mean that Hort, in

spite of the limitations of genealogy, was essentially on

the right track. Rather than expressing alarm, as does Aland, that in spite of so many advances in the past 75 years our resultant text is "Hortian," perhaps we should express wonder that Hort, without our discoveries and advances, reveals such balanced judgment. 3. The analysis of P66 and F75 in this study has

Indicated that another major cause of unrest over the current methodalthough it is again unrest over the results rather than the method itselfhas been removed, namely the recensional nature of the Neutral texttype. The combined

witness of these two MSS suggests that the Neutral texttype as a recension, if not impossible, is at least improbable. 4. The conclusion of this study as to the "rela-

tively pure line of descent" of the Neutral MSS is not a return to genealogy. It is rather an attempt to evaluate

the place of this textual tradition as "external evidence" in the eclectic method. The fact that all analyses of this

textual tradition have revealed a number of secondary readings means that one cannot simply follow the Neutral text wherever it leads. There must be fully as much emphasis placed on "relatively" as on "purej" but, by the same measure, one should be wary of rejecting readings of this texttype because they are allegedly "recensional." 5 The place of the Neutral textual tradition, as it

has been evaluated in this study, in the eclectic method as a whole may be illustrated from the following general

267
guidelines: (it should be noted that these re "Illustra tions" of the method, not necessarily rules always to be followed.) A. The full combination of the best of external

evidence, i.e. the Neutral KSS plus other early and widely distributed witnesses/ and the best of internal evidence, i.e. the reading which best explains the others, is to be preferred. This Judgment, made by Hort, continues to find

general acceptance. B. The combination of Neutral witnesses, with or

without other support, and the canon ardua lectio potior is usually to be preferred. An example of this is the argument given above (p. 243) for the word order ot in John 12:l6. C. The reading of a good Neutral KS, which has

secondary Neutral support and other non-aligned support, when that reading best explains the others, is to be preferred. An example of this is the omission in John 9:38-39 of 6* * ntoreuas, uptc* mxt . xat etnev by P75 * V b 1 bo^. Besides the com bination of this external evidence, the following arguments support the omission as original: (l) Piety would add such

words; there is no explanation for their omission, not even by ecribal error. (2) The expression oc c p is nontn

Johannine. Both the use of (c for the continuation of

268
narrative and the verb *\ are rare in John; their combina t i o n i s therefore doubly peculiar. (3) These words f i t neither the immediate context of John nor the greater context, where t h i s type of worship cocoes a t the cliaax of the Gospel in the words of Thomas. D. When a l l other considerations are equal, the

reading of the Neutral texttype, especially when i t has other non-aligned support, i s to be preferred, principally on the basis of the "relative purity" of t h i s textual tradition. This i s e s p e c i a l l y true a t insignificant* points of variation; but i t nay a l s o be true at such varia tions as the "seventy or seventy-two" i n lata* 10:1 and 17E. Vhen there i s good non-Neutral XS support for

the ardua l e c t i o and i t best explains the reading found i n the Neutral KSS, heavy emphasis should be put on the r e l a t i v e "purity" of the Neutral tradition. the Neutral t r a d i t i o n i s probably secondary. An example here i s the & of R D pc i t bo sy 6 in John 7:8, which i s t o be preferred to the ovro* of the majority. F. A reading which best f i t s an author's s t y l e , or In other words,

the context, and which best explains the other readings, w i l l often be preferred against strong external a t t e s t a t i o n . When such occurs, however, the internal arguments DEust be particularly strong. Some of the Lucan "Western non-inter polations" s t i l l seem to be preferred on t h i s canon.

269
Another example at this point is the reading of ?5rld K # 77 218 b e ff2
Z7c.

(a

) against

the of the rest in John 1:34. This reading seems to be preferred for the following reasons: (1) The change, whichever direction it took, ie almost certainly of doctrinal motivation. It seems sore likely that "orthodoxy" altered what say have been a basis for heretical argument, than that "heretics" should alter the text only at this one point in the Gospelunless, of course, "heresy" considered this confession to be important only at Jesus1 baptism. (2) The author himself uses a full complement of Christologlcal titles in his homologia. In the mouths of different witnesses are the titles (1:29), (l:4l), (1:49), (4:42),

(6:14), (6:69)* (11:27) Two things should be noted about these confes sions. First, they are always appropriate to the situation and to the people making the confession. Secondly, it will be noted that no one of these is ever repeated as confes sions in the Gospel except at 11:27 (which is also the basic hoaologlon of the Gospel [20:31])It is therefore

quite likely that the author himself used at this appropriate place in the Gospel. Whereas the other confessions, although soon to pass out of the church's

270 i confessional life, were not Christologically difficult, * had clear Ebionite Implications and therefore waa probably changed in the interest of John's basic confession. -^ |

Finally, two words of caution should be raised to the eclectic critic who has "Hortian" tendencies, i.e., the tendency to adopt a Neutral reading on the basis of a greater emphasis on KS evidence than on internal questions. 1. One must be aware of the problem of circularity in his argument. This is the problem of arguing for the

Keutral KSS as "best" on the basis of internal arguments at given readings, and then arguing for the same readings as "preferredn because they have the "best" support. 2. One must be aware of the danger of prejudging a

reading on the basis of preference for certain KSS. The fact that all KSS have some errors in their text means that one must hear all the evidence before a choice is made. This brief presentation of some guidelines for "reasoned" eclecticism does not offer a new method; nor has that been the intention of this study. The purpose here

5*vernon H. Neufeld (The Earliest Christian Confes sions Grand Rapids, 1963 ) dismisses this probably Johannlne confession in a footnote with the words: "The meaning, however, is practically the same" (p. 73, n. 2 ) . To the author of the Gospel this may have been true, but it over looks the clearly doctrinal implications of the variant.

271

has been to demonstrate that the recent papyrus discoveries, P66 and P75# tend to support "reasoned" eclecticism as a valid method. Colwell's observation of the current situa tion seems to be true: "The best New Testament must be Rather than search for a new

chosen verse by verse."55

method, the present methodological task would seem to be the implementation and refinement of the current, valid method.

5^What is the Best New Testament? (Chicago, 1952), P. 72.

APPENDIX I VARIATION-UNITS AND SINGULAR READINGS IN JOHN 4 Variation-units in John 4. The following list includes only those units of variation where at least two of the early MSS agree against the rest. Therefore, only the MSS involved in the discussion in Chapter II are cited, along with the TR. The top reading in each instance is

that of Nestle-Aland (25th edition). 4:1 4:1 4:2 4:3 4:5 4:5 4:9 P66c P75 A C W TR ?66* D p66 75 * C D TR * V 66 P75 C W TR D

?66 75 C D W TR omit A 75 A TR P66C D W

66 P75 A C D W TR Iou6atot P66 P75 A C W TR omit K* D omit exetvn P75 P66 A C D W TR K*

4:11

4:11

P66 P75 A C TR W D 273

4:12 4:14 4:14 4:14 4:15 4:15

A D TR P66 P75 C 6 ^ * ?66 75 A C () TR * D (C lac) '

75 A D 66 V TR

66 75 A TR cyoe > D V *

P66c 75 A C TR P66* D P66 75 A C D V TR P66 P75 C* * R c D TR (tt* lac)

cpxcopat omit

4:16

4:17

A D TR P66 (P75) C

4:17 4:17 4:21 4:23 4:24 4:24 txe

66 75 TR C* D

66 75 A C TR & D 66 75 C* A (D) TR P66c P75 A C D TR 6* * * ] * > 66 75 A C TR

66 75 A C TR * D

4:25 4:25 anavra

?66 75 A C TR * D

P66 75 C* A D TR

275
4:27 4:27 4:29 4:30 tnt four P66 P75 A C V ffl
y T O U T K* D

tiev tnev

P66 P75 A C TO D

C* 66 75 A D TO 75 66 W TR C D

xat ev ev bt xat CY note crt omit

4:31

66 C* D P75 A TO V

4:34 4:35 4:36 4:37 4:38 4:38 4:39 4:42

A TO 66 75 C D 66 A C TO P75 D ? 75 C A D 51 (75 l a c )

<mtpanr xat OTTctpw onearaXxa a re 6e xat ort omit

C* 66 TO

66 75 A C TO D

P66 P75 C TO D* P75 C* 66 TO P75 A C RC TO P66 D K* P66 P75 A C D TO P66 A C TO 75 * D

4:42 4:42

XaXtav

4:42

4:42 4:45 4:45 4:46 4:47 4:47 4:50 4:50

66 75 C V TR D nap omit P66 P75 C V A D TR ore 66 75 A C V TR * D

66 75 A C W * D TR * ^ 6c * 66 75 A C V TR D P66 c P75 A C D V TR 6* F66 P75 A D * TR * C P66 P75 D W A C TR

eiTtoTcuoeY xai errtOTeuoer

4:51 4:51 4:51

or d i r e r avre (75) A C c ctner aim 66 D W TR * o r Kc ot (66 75 A C TR) (W) D 66 75 A C W TR

omit P75 66 A C TR t D t omit 66 75 A C V TR D 66* 75 R A C W P66c D TR

4:51 4:51

4:52 trap 66 A C D W TR F75

277 4:52 etnav at etnav P66 P75 C W A D TO

etnav
4:53

?66*
75 A D W TO 66 C

4:53 4:53 4:54 4:54

cxtvn P75 K* C ev exctvq p66 A D TO 66 P75 A b ort D W TO 6e P66 P75 C* W

TOUTO A D TO

oevrepov tito\r\acv 6eurepov enotnacv

P66 A C D TO it W 75

Singular (and sub-singular) readings i n John 4 . TOe f i r s t iteias l i s t e d for each M are singular readings; S the second items are variants where the given M has S versional support. A. Papyrus Bodmer I I (P66)
4:6 4:6 4:9 4:12 4:14 . txadiJtTO 1. 1. (cum 124l)

netv 1. netv om. ot ante ev aurco 1. ev

Papyrus Bodmer XV (P75) 4:17 4:18 4:37 4:41 1. etnev . omit verse nXetov 1.

4:47 4:54

. ocurcpov 1. 6curepov enotncrev

Codex Vaticanus (B) 4:40 4:52 1. , . (cum )

Codex Sinaitlcus () 4:11 4:12 4:12 4:14 4:15 4:17 4:19 4:20 4:24 4:42 4:46 4:47 4:53 4:7 exetvn 1. (or omit) 1. xat 1. xat om 2 >& 1 cvda5c om. xat etnev om. xupte om. (cum 245) (cum 348)

l. xat aXnSeta xat 1. TC . add post 0 om. add ante (1187 b J r 1 sy 8 0 )

4:18 . (c f q r 1 vg) 4:27 ennXeav 1. tjXOav (e J q r 1 )

4:33 Xeyowtv 1. (b r 1 ) 4:39 om. *? (482 a e)

279 4:40 4:42 1. (sa bo arm) 1. (sa bo s y 0 ) 4:45 ot 1. (a b f sa bo) 4:45 4:47 4:49 4:50 1. (it vg) om. (sy) 1. (b d e ff2 1 q r 1 ) 1. () ctnov (ea bo aye) Codex Bezae (D) 4:11 4:12 4:14 4:33 4:36 4:38 4:45 v 1. (cum 49 91) om. post om. 1. 1. 1. 1. (cum 182) bis

4:46, 48 1. 4:51

1. ot

4:9

> 1. > ( i t ey30 sa)

4:9 4:11

om. (J arm) 1. ( s y 8 , c bo)

280 4:19 4:21 4:28 om. (a b ff 2 J 1 r 1 ) T O opet 1 opet (a b r^ ey 8 ) O \ 1. (b ff2 l q r* sa ana)

4:29 4:33 4:36 . (q) . ,(a b q r 1 ) xat 1. xat (ey 8 0 5 arm aeth) 4:37 coTtv 1. (301 482 aur a b c f ff 2 1 q) 4:45 4:51 P. om. ev (e) add post * (b)

Codex Alexandrlnus (A) 4:26 om. ante

Q. Codex Ephraemi rescriptus (C) 4:2 4:40 4:47 . on. 1. nap . xat

Codex Washlngtonianus (W) 4:11 xat 1. 4:11 4:11 4:12 4:14 4:27 xat 1. 1. 2 add TO Jarv post add post add post

281 4:31 4:45 4:47 4:48 4:51 \ cv 1. or ev 6c om. aurot nxev 1. * om. ante cirrou 1. ot aura 4:6 4:7 4:23 4:29 4:52 add 6c post (b) add ante cpxcrai (a b e ff 2 l q r add cv post (a b) om. pot (ff2)
1

sy30)

. (a b sa)

APPENDIX U CORRECTIONS OP P66 WHERE THE ORIdKAL TSCT (P66*) HAS SIHGULAR OR SUB-SINGULAR RiADlKGS In the following list of readings only those correc tions are included in which the readings of both P66* and P66c are grammatically and contextually possible. In each case the reading on the left is that of P66*; the reading on the right is that of P66c. Any other witnesses which agree with P66* follow the first reading in parentheses. 1:49 2:2 2:13 2:15 2:20 2:25 3:2 3:19
3:21

pappt, t (1241)

] - ** ] + at

6c (440 579 1071 aur f J l q ) 6e *at 3 - * 3

xat ev Tptotv

] xat Tptotv

nept (bo**) 3 nept avepemoo (L 239) (472) 3

notarr 3

3 eav

3*31 3:31 4:6 4:6

x eoTtv

3 + xat ex 3

tx xa0teTO

3 exa8ecTO

ent (1241) 3 tm

282

4:9 4:12
4:42

nttr

3 nctr

3 utot
(053 1 87) 3

5:2 5:2 5:5 5:6 5:22 5:36 5:43 6:1 6:9

6:42 6:52 6:60 6:64 7:18 7:28 7:37 7:44 8:28 8:40 8:46 8:48 9:8 9:10

3 3 " % xet 3 n1 X* (l24l) 3 * 3 aura *T 3 - o * 3 + & (473 ) 3 Tt

OTt 3 3 + **

3 + * 3 3 3

(sa sy 5 0 aeth) *

(131) 3 pot 3 3 3 + pot

(71 249 251) 3 urewxtnaar 3 - 3 ^

284 9*30 9x39 10:4 10:7 10:33 10:38 11:2 xat etnev 3 xat etncv 3 +

(1071 1241 1293) ctnev ()

3 cx0aXi)

3 + ] - 3 * 3

ntoTcuotjTe (1200 1334)

cat no evet

11:3

cneoTetXev Kapta 3

11:5 11:7

3 + 3

11:21 11:34

3 3 xat xbe ] + noXtv 3

t6e

11:54 12:1 12:2 12:2 12:3 12:7

(13 122 bo) 3 + 3 + 3

3 ] ] eoTat ] 3 - xat

12:16 12:26 coTtv 12:37

12:45 eewpet xat

13:15 (472 700) ] + 13:16 3 +

285 13:20 14:2 ctnov ] -

3 ctnov av 3 3 * * "

14:23

14:23 etoeXeuoope*a 15:3 ev uptv ] - cv

15:10

15:10 (e) 3 15:16 15:19

)

3 + 3

15:22 15:25 16:7 17:11

3 +

3 + 3 + cav 6c ^ * 3

17:11 17:12 17:12 17:19 18:2 18:5 18:12

3 + tva oootv ev 3

3 3 +

w>tv (544 a b c ) 3 +

3 + 3 3 +

18:15

19:14 3 + 3 +

19:17 3[] [3 19:28

3 + tva

APPENDIX III DISAGREEMENTS BETWEEN P75 AND IN LUKE ; The following list of disagreements includes only

those variants which are grammatically and contextually possible. Excluded, therefore, are such "nonsense" readings as toxuvl. in 11:11, where the following clause makes the toxuv contextually impossible, and the epxerat 1. in 23:29* where the plural subject makes impossible. Also excluded are purely orthographical disagree mentsalthough differences in the spelling of proper names are includedand readings where the text of P75 is purely conjectural. Comments on some of these excluded conjec tural readings follow the main list. In each case the reading on the left is that of P75 and the reading on the right that of B. 3:20 3:22 4:35 4:35 4:41 5*1 5:2 [ ][]
e e

(C Corrector)

] -

] ] ] ] ] 6 ] TiXota 286

? ? "

287 6:3 6:25 6:28 6:38 6:41 6:49 7:6 7:22 8:5 8:7 8:9 8:13 8:20 8:21 8:22 8:22 8:27 8:28 8:29 9:1 9:9 9:12 9:18 9:28 9:28 9:29

] 2 3 -

3 3 C ,

][][ 3 rt 3 - ?

3 C -xat 3 -*

3 3 3 C

ouTot xat

3 3 + 3

3 +

3 3

3 ute 3 3

] + 3 3 3

[ ][ 3 -

[ 3

288 9:3^ 9*39 9:42 9:48 9:^8 9:50 9:58 9:59 9:59 9:62 9:62 10:18 3 +

] 3 3

[ ] ] ] ] 3 C+

] 3 3 - ev 3 *

ev

10:19 10:24 10:27 10:31 10:37 10:39 10:39 10:39 10:40 10:42 10:42 11:2 11:7

3 +

3 + 3 - C+

3 - 3 3 3
+

C+ C-

3 3

3 -

3 - 3 -

11:11 11:13

TTtrrepa

] 0 C 3 3
+

11:15, 18, 19 11:18 11:19 11:22 11:22 11:23 11:24 11:25 11:27 11:30 11:31 11:31 11:33 11:36 11:39 11:39 11:41 11:42 11:42 11:44 11:48 11:50 12:8 12:18 *

3 0 3

3 3 oxopntjet ] - 6e 3 +

6e

3 + 3 C + 3
+

PaatXtaoa

3 3

3 ev 3 3 3

3 3 C + 3 ot

3 3 3 C - 3 -

12:22 12:24 12:28 12:29 12:29 12:31 12:39 12:39 12:42 12:43 12:45 12:45 12:48 12:53 12:53 12:56 12:58 13:1 13:1 13:2 13:5 13:7 13=7 13:11 13:13 13:14

oeport

] ]

3 et 3

3 xat 3 + ] + -

C+

3 xat ] 3

notouvTa 3 3 + 3

3 - 3
+

C+

xat 3 - 3 + >

3 3 3 3 - C

^
rt v

3 -

3 C + 3 - 3 +

13:1* 13:15 13:17 13:21 13:25 13:27 13:29 13:30 13:31 13:32 13:33 13:3^ 13:35 14:8 14:8 14:10

ev

3 - C + 3 C ] 3 ] 3

3 * cmo 3 - 3

3 + 3

3 + voootav 3

Ue

3 + 3 +

3 3 not

14:13 14:17

3 3 * 3 -

14:21

14:23 C - 14:25 14:26 3 + ) 3 + 3 8t vat

14:26 14:27 14:28

3 3 3

14:28

14:32 14:34 14:35 15:4 15:6 15:10 15:12 1513 15:17 15:18 15:21 15:22 15:22 15:22 15:24 15:24 15:30 16:1 16:4 16:12 16:15 16:15 16:17 6: 1:19 16:22

eav

] + ] + ] - 3 C ] ] - "

] 6 3 ] ]

] + ( ] 3 ] - ] ]

3 + ] G

] 3 ] C

] 3 3 3 3 -

ptav

293 16:22 :27 16:27 16:30 16:31 17:1 17:6 17:7 17:12 17:22 17:23 17:30 17:34 18:10 22:9 22:19 22:24 22:26 22:39 22:40 22:41 22:47 22:50 ] + 6c ] etnev

] ] ] 3;+

C +

3 +'

3 ] ] C ] C

t6ou 3 ] ]

] +

] - C + ] + C ] - c + ] - C + ] ] ] ]

3 3 3

22:51 22:55

22:61

23:2 23:3 23:5 23:6 23:8 23:11 23:11 23:11 23:12 23:12 23:23 23:25 23:26 23:29 23:29 23:31 23:35

] ] + OUTUU ] ] C +

] ] - ] ] ] ] ] ] -

ort

3 + ] at xoiXtat 3

TU)

C ]

23:39 23:45 23:46 23:50 23:50 23:53 24:1 24:15

] C ]

] 3 - xat ] -

] ] ] C

Kat

295 24:17 24:18 etnev 3 + 6c C [sic] 3

24:21 24:26 24:27 ] C ] 3

24:28 24:39 24:39 24:39 24:44 24:47 24:47 24:49 24:49 24:52 24:53

] 3 + 3 + 3 3

ev > 3 *<** eym 3 3 xat > 3

3 - C +

3 +

The following variations from B are found in the editlo prlnceps, but they have been left out of the fore going list as not demonstrable from the extant text: 3:36 - 3 + This is possible, but

the conjecture probably stems from the fact that D omits this name. It is true that "lacuna longius esset," but the missing name could as easily be or , cr some

295
combination. 6:22 - 3 This seems a likely conjec

ture, but the lacuna is again of such nature to allow otber possibilities. 6:23 ] 3 . This read

ing is purely conjecture.* I submit that the text should read ]<> for the following reasons: (l)

What Kartin and Kasser read as an in rwi looks more like t this scribe's than his w. (2) The scribe frequently has

an uncorrected text consisting of the omission of a single letter, but he never spells out the iota subscript. (3)

Tne lacuna following seems too long to support only three letters, but would easily support five without crowding. 6:39 Kartin and Kasser conjecture that P75 omits

, but this Is highly doubtful. What they read as ou* seems almost certainly to be the c[o]y of . (Note that the scribe's Is always of such size that it would be almost certainly missing here.) Moreover, the lacuna supports , but Is Inexplicable without It. (Note how much smaller his letters are in this line than in those above.) 6:46 rt] 3 + 6e. There seems to be no good

reason to conjecture an omission in P75 at this point. If the left margin Is even vaguely straight, the lacuna Is well able to support four letters.

297
7:15 ] ] cxoOtoev. The lacuna here is of such nature that P75 could as easily support as not. 9:1 } ] . The lacuna does not seem

long enough to support the necessary letters of the ordinary text here, but what might be missing is purely conjectural. 9:9 # ] . Since this scribe's

is so small, a conjectured omission where the lacuna could possibly support an seems doubtful. 9:27 ] ] - on. This is perhaps a

good conjecture based on the length of the lacuna, but one wonders what might have been conjectured had D not read o n here. 9:28 xat ]0> ] - . Again the con

jecture is based on the length of a lacuna at the end of a line. The reading of P75 here could go either way. 10:1 [ ava ] - . The editio

princeps is clearly wrong here. What is read as ay should be the cy of . See K. Aland, "Neue Neutestamentliche Papyri II,n NTS, XI (1964/5), 7. 22:17 ]}[ ] -

The editors say that "sine lacunam non explet." But that does not seem to be quite true. It is only the difference of fifteen and seventeen letters, and I count a number of places on this page where fifteen letters would fit in this lacuna.

BIBLIOGRAPHY

BIBLIOGRAPHY A. TEXTS, COLLATIONS, AND CRITICAL EDITIONS

Barnard, P. Mordaunt. The Biblical Text of Clement of Alexandria In the Four Gospels and, the Acta of the ApostleiT Cambridge: At the University Press, 199. Vol. V, No. 5, Texts and Studies. Bibllorum Sacrorum Graecus Codex Vaticanus. Rome: Congregationls de Propaganda Fide, looo. Bover, Jose M. Novl Testamentl Biblla Graeca et Latlna. Madrid: Consejo Superior de Investigacionee Cientificas, 19^3. Brooke, A. E. The Commentary of Qrigen on S. John's Gospel. 2 vols. Cambridge: At the University Press, Burkitt, F. Crawford. Evangelion da-Mepharreshe: the Curetonlan Version of the Four Gospels, with the Readings of the Slnai-Pallmpsest. 2 vols. Cambridge: At the University Prese, 1904. Cronin, H. S. Codex Purpureus Petropolltanus (N). Cambridge: At the University Press, 1&99.Vol. V, No. 4, Texts and Studies. Facsimile of the Codex Alexandrlnus: New Testament and Clementine Epistles. London: 1879. Geerlings, Jacob. Family 13 (The Ferrar Group), The Text According to John. Salt Lake City: University of Utah Press, 19o2. Studies and Documents, No. 21. . Family 13 (The Ferrar Group). The Text According Eo Luke. Salt Lake City: University of Utah Press, I96I. Studies and Documents. No. 20. [Horner, George.] The Coptic Version of the New Testament in the Northern Dialect. Vol. II. The Gospels of S. Luke and S. John. Oxford: At the Clarendon Press, 299

300
[Horner, George.] The Coptic Version of the New Testament In the Southern Dialect. Vol. II, The Gospel of S7 Luke, Vol. Ill, The Gospel of S. Johru Oxford: 5F the Clarendon Press, 1911. " jQllcher, Adolf. Itala: das Neue Testament In altlatelnlscher Uberlieferung, ed. W. tetzkow and Kurt Aland. Band III, Lucas-evange Hum, 195^ Band IV, Johannesevangel ium, 1963. Berlin: Walter de Gruyter. Kenyon, Frederic G. The Chester Beatty Biblical Papyri: Descriptions and Texts of Twelve Manuscripts on Papyrus of the Greek Bible. Fasciculus I, General Introduction. Fasciculus II, The Gospels and Acts, Text and Plate3. London: Emory Walker Limited, 1933. Lake, Kirsopp. Codex 1 of the Gospels and Its Allies. Cambridge: At the University Press, 1902. Vol. VII, No. 3# Texts arid Studies. , and Helen Lake, eds. Codex Slnaiticus Fetropolitanus. Oxford: At the Clarendon Press, 1911 , and Silva New, eds. Six Collations of New 3>e"stament Manuscripts. Cambridge : Harvard University Press, 1932. Harvard Theological Studies, No. XVII. Martin, Victor, ed. Papyrus Bodmer II: Evangile de Jean, chap. 1-14. Cologny-Geneve, Switzerland: Bibliotheque Bodmer, 1956. . Papyrus Bodmer II, Supplement: Evangile de Jean, : chap. 14-21. Cologny-Geneve, Switzerland: Bibliotheque Bodmer, 1958. , and J. W.,B. Barns, eds. Papyrus Bodmer II, Supplement: Evangile de Jean, chap. 14-21. Nouvelle edition augmented et corrlgee. Cologny-Geneve, Switzerland: Bibliotheque Bodmer, 1962. ,_, and Rodolphe Kasser, eds. Papyrus Bodmer XIV-XV: Evanglies de Luc et Jean. 2 vols. Cologny-Geneve, Switzerland: Bibliotheque Bodmer, 1961. Nestle, Eberhard. Novum Testamentum Graece. 25. Auflage, ed. Erwln Nestle and Kurt Aland. Stuttgart: Wurttembergische Bibelanstalt, 1963.

301
Sanders, Henry A. The Washington Manuscript of the Four gospels. New York: The .MacMilian Company, 1^12. Part I of The New Testament Manuscripts In the Freer Collection. Schmidtke, Alfred. Die Evangellen eines alten Uncialcodex (BK -Text) nach elner Apschrlft des drelzehnten Jahrhunderts herausgegeoen. Leipzig, 1903. Scrivener, Frederick H. Bezae Codex Cantabrlglensls, Being an Exact Copy JIn Ordinary Type^ Cambridge, 1864. . A Full Collation of the Codex Slnaltlcus with the Received Text of the New Testament. Cambridge, 1867. Tasker, R. V. G., ed. The Greek New Testarsent, Being the Text Translated in the New English Bible, 1961. Oxford University Press and Cambridge University Press, 1964. Thompson, Henry Francis Herbert, ed. The Gospel of St. John According to the Earliest Cop~tic Manuscript."" London: British School of Archaeology in Egypt, 1924. Tiechendorf, Constant in. Novua Testaraentxm Graece, 8th ed. 3 vols. Leipzig; Giesecke and Devrient, 1869. von Soden, Hermann Frelherr. Die Schriften des Neuen Testaments. 2 vols, in 4~ Gottingen: Vandenhoeck und Ruprecht, 1913 Vestcott, Brooke Foss and Fenton John Anthony Hort. The New Testament in the Original Greek, 2nd ed. 2 vols. London: MacMlllan and Co., Liaited, I896. B. BOOKS AND ARTICLES

Abbott, Edwin A. Johannlne Grassnar. London: Adam and Charles Black, 1906. Aland, Kurt. "Das Johannes-evangeliuni auf Papyrus," Forschungen und Fortschrltte, XXXI (1957), 50-55 . Kurzgefasste Llste der griechlschen Handschriften des Neuen Testaments. Berlin: "Walter de Gruyter, . MNeue Neutestaraentliche Papyri II,B New Testament Studies, IX (1962/63), 303-316; X (1963/64), 62-79-

302 Aland, Kurt. "Papyrus Bodmer II, ein erster Berlcht," Theologlsche Llteraturzeitung, LXXXII (1957), 161-184. . "The Present Position of New Testament Textual Criticism," Studia Evangellca, ed. K. Aland, et al. Berlin: Akademie-Verlag, 1959. Pp. 717-731. . "The Significance of the Papyri for Progress in Rew Testament Research,1"' The Bible in Modern Scholarship, ed. J. Philip Hyatt"]! Nashville: Abingdon Press, 1965. Pp. 325-3^6. Barns, J. W. B. "Papyrus Bodmer II, Some Corrections and Remarks," MuseOn, LXXV (1962), 327-329. Barrett, Charles Kingsley. The Gospel According to St John. London: S.P.C.K., 1955. . "Papyrus Bodmer II, A preliainary report," Expository Times, LXVTII (1956/57), 174-177. Bernard, J. H. A Critical and Exegetical Commentary on the Gospel According to St. John. 2 vols. Edinburgh; T. & T. Clark, 192b. (International Critical Commentary) Blrdsall, J. Neville. The Bodmer Papyrus of the Gospel of John. London: The Tyndale Press, 190O. The Tyrdale New Testament Lecture, July, 1953. . "Current Trends and Present Tasks in New Testaaent Textual Criticism," Baptist Quarterly, XVTI (1957), 109-114. . "The Text of the Fourth Gospel, Some Current Questions,u The Evangelical Quarterly, XXIX (1957), 195-205. Blass, Priedrich. Philology of the Gospels. London: KacMillan and Co., Limited, IbSs. , and Albert Debrunner. A Greek Grammar of the New Testament. Tenth German edition, trans, and rev. Robert W. Funk. Chicago: The university of Chicago Press, 1961. Boieinard, M.-E. n A propos de Jean V, 39. Essai de critique textuelle," Revue Blblique, LX (19^), 5-34. . "Critique textuelle et citations patristiques," Jtevue Blblique, LVTI (1950), 3SS-408.

303 Bolsnard, M.-E. "Dane l e s e i n du Pere ( J o . , I , 1 8 ) , " Revue Bibllque, UX (1952), 23-39 . "Lectio b r e v i o r , p o t i o r , " Revue Biblique, LVIII 951)/ 161-168. . "Le papyrus Bodmer I I , " Revue Biblique, LXIV (1957), 363-398.
. "Problemes de critique textuelle concernant le Cuatrieme Evanglle,1* Revue Biblique, IX (1953), 347-371. Botte, Bernard. "Le texte du quatrieme Evangile et le Papyrus Bodmer II," Bible et Vie Chretienne, XXIV (1958), 96-107. Bousset, Wilhelm. Textkritische Studien gum Neuen Testaaent. Leipzig: J. C. Hinrichs, lt>94. Breun, Francois-Karie. " nouveau papyrus Johannique," Revue Thomlste, LVII (1957), 79-84. Brinksann, B. "Eine neue Papyrus-Handschrift des Johannesevangeliums aus dem 2. Jahrhunderts," Scholastix, (1957), 399-410. Burton, Ernest deWitt. Syntax of the Moods and Tenses in Key Testament Greek. Fourth edition. Chicago: The university of Chicago Press, 1900. Cadbury, Henry J. "A Proper Name for Dives," Journal of Biblical Literature, LXXXI (I962), 399-402": . "The Naae for Dives." Journal of Biblical literature, LXXXIV (1965), 73T Clark, Albert C. The Act3 of the Apostles. Oxford: At the Clarendon Press, 1933. . The Primitive Text of the Gospel3 and Acts. Oxford! At the Clarendon Press, 1914. Clark, Kenneth W. nTne Effect of Recent Textual Criticism upon New Testament Studies," The Background of the New Testament and Its Eschatology, ed. W. D. Davies and D. Baube. Cambridge: At the university Press, 1954. Pp. 27-51.

304
Clark, Kenneth W. "The Manuscripts of the Greek New Testament," New Testament Manuscript Studies, ed. K. M. Parvls and A. P. Wikgren. Chicago: University of Chicago Press, 1950. Pp. 1-24. . "The Text of the Gospel of John in Third-Century Egypt," Novum Testaroentum, V (1962), 17-24. . "The Theological Relevance of Textual Variation in Current Criticism of the Greek New Testament," Journal of Biblical Literature, LXXXV (1966), 1-16. Collins, John J. "Papyrus Bodmer II," Catholic Biblical Quarterly, XX (1958), 281-289. Colwell, Ernest Cadman. "Biblical Criticism; Lower and Higher," Journal of Biblical Literature, LXVTI

(1943),

tt2~.

. "The Complex Character of the Late Byzantine Text of the Gospels," Journal of Biblical Literature, LIV (1935), 211-221. . BA Definite Rule for the Use of the Article in the Greek New Testament," Journal of Biblical Literature, LII (1933), 12^20"! . "Genealogical Method: Its Achievements and Its Limitations," Journal of Biblical Literature, LXVT (1947), 109-13J: . "Method in Locating a Newly-Discovered Manuscript within the Manuscript Tradition of the Greek New Testaoent," Studia Evang;elica, ed. X. Aland et al. Berlin: Akademie-Verlag, 1959- PP 757-777. . "The Origin of Texttypes of New Testament Manuscripts," Early Christian Origins, ed. A. Wikgren. Chicago: Quadrangle Books, 1961. Pp. 128-138. . "Scribal Habits in Early Papyri: A Study in the Corruption of the Text," The Bible in Modern Scholarship, ed. J. Philip Hyatt. Nashville: Abingdon Fress, 1965. Pp. 370-389. . "The Significance of Grouping of New Testament Manuscripts," New Testament Studies, IV (1957/58), 73-92. . What Is the Best New Testament? University of Chicago Press, 1947. Chicago:

305
Colwell, Ernest Cadman, and Ernest W. Tune. "The Quantitative Relationships Between MS Text-types," Biblical and Patristic Studies in Memory of Robert Pierce Casey, ed. J. N. Blrdsail and R. W. Thomson. Freiburg: Herder, 1963. Pp. 25-32. , and Ernest W. Tune. "Variant Readings: Classification and Use," Journal of Biblical Literature, LXXXIII (1964), 253-261. Dover, Kenneth J. Greek Word Order. .Cambridge: The university Press, I960. Duplacy, Jean. Ou en est la critique textuelle du Nouveau Testament? Paris: J. Gabalda et Cie, Editeurs, 1959 Epp, Eldon J. "Some Important Textual Studies," Journal of Biblical Literature, LXXXIV (1965), 172-175. Fee, Gordon D. "The Corrections of Papyrus Bodmer II and Early Textual Transmission," Novum Testamentum, VII (1965), 247-257. "Corrections of Papyrus Bodmer II and the Nestle "Greek Testament," Journal of Biblical Literature, LXXXIV (1965), 66-72: Peine, Paul, and Johannes Behm. Introduction to the New Testament. Fourteenth revised edition, ed. Werner Georg Kummel, trans. A. J. Mattill, Jr. Nashville: Abingdon Press, 1966. Filson, Floyd V. "More Bodmer Papyri," Biblical Archaeologist, XXV (1962), 50-57. . "A New Papyrus Manuscript of the Gospel of John," Biblical Archaeologist, XX (1957), 53-63. Fitzmyer, Joseph A. "Papyrus Bodmer XIV: Some Features of Our Oldest Text of Luke," Catholic Biblical Quarterly, xxiv (1962), 170-179. Funk, Robert . "Papyrus Bodmer II and John 8:25," Harvard Theological Review, LI (1958), 95-100. Grant, Frederick C. "The Greek Text of the New Testament," An Introduction to the Revised Standard Version of the New Testament. The International Council of Religious Education, 1946. Pp. 37-^3.

306
Oreenlee, J. Harold. "A Corrected Collation of Codex Zacynthius (Cod )," Journal of Biblical Literature, LXXVI (1957) 237-241: . Introduction to New Testament Textual Criticism. Srand Rapids: William B ^ Eerdraans Publishing Company, 1964. Gregory, Caspar Rene. Canon and Text of the New Testament. Edinburgh: T. & T. Clark, 1907,. Harris, J. Rendel. "Some Speculations of Textual Criticism Justified," Side-Lights on New Testament Research. London: The Kingsgate Press, n.d. (Lectures, 190). Pp. 79-110. Hartke, W. "Bemerkungen zu Papyrus Bodmer II," Forschungen und Fortschrltte, XXXII (1959)* 122-126. Hills, Edward F. "The Inter-relationship of the Caesarean Manuscripts," Journal of Biblical Literature, LXVIII (1949), 141-159. Hoskier, H. C. Codex and Its Allies, A Study and an Indictment. 2 vols, in I. London: Bernard Quaritch, 1914. Huffman, Norman. "Suggestions from the Gospel of Jferk for a New Textual Theory," Journal of Biblical Literature, LVI (1937), 347-359. Hunger, Herbert. "Zur Datierung des Papyrus Bodmer II (P66)," Anzelger der bsterreichischen Akademle der Wlssenschaften, philosophisch-hlstorische iQasse, I960, Nr. 4, pp. 12-23. Jeremias, Joachim. Philadelphia: The Lord's Prayer, trans. John Reumann. Fortress Press, 164.

Karnetzki, Manfred. "Textgeschichte als Uberlieferungsgeschichte," Zeltschrlft fur die Neutestamentliche Wlssenschaft. XLVII (195b). 170-10. Kenyon, Frederic G. "Hesychius and the Text of the New Testament," Memorial Lagrange. Uppsala: Seminarium neotestamentium upsaliense, 1938. Pp. 245-250. . The Text of the Greek Bible. New edition. London: Gerald Duckworth and Company, Limited, 1949.

307
Kilpatrick, George D. "Atticism and the Text of the New Testament," Neute3tamentliche Auf3atze, ed. J. BInzler. Regensburg: Verlag Friedrlch Pustet, 1963. Pp. 125-137. . "An Eclectic Study of the Text of Acts." Biblical and Patristic Studies in Memory of Robert Pierce Casey, ed. J. N. Blrdsall and R. W. Thomson. Freiburg: Herder, 1963. Pp. 64-77. ' ' : . "Some Notes on Johannlne Usage," Translator, XI (i960), 1-5. The Bible

. "Some Notes on Marcan Usage," The Bible Translator, VII (1956), 2-9, 51-56, USSl . "The Transmission of the New Testament and Its Reliability," Synopsis of a lecture delivered before the Victoria Institute, April 15, 1957. Pp. 92-101. . "Western Text and Original Text in the Epistles," Journal of Theological Studies, XLV (1944), 6-65.

<

. "Western Text and Original Text in the Gospels and Acts," Journal of Theological Studies, XLIV (1943), 24-3o~;

Kim, K. W. "Origen's Text of John in His On Prayer, Com mentary on Matthew, and Against Celsus," Journal of Theological Studies, n.s. I U950J, 7^84. King, Merchant A. "Notes on the Bodmer Manuscript of Luke," Blbliotheca Sacra, CXXII (1965), 234-240. , and Richard Patterson. "Textual Studies in the Bodmer Manuscript of John," Blbliotheca Sacra, CXVII (I960), 164-177, 258-266. Klijn, A. P. J. "Papyrus Bodmer II (John i-xiv) and the Text of Egypt," New Testament Studies, III (1956/57), 327-334. Kubo, Sakae. P72 and the Codex Vaticanus. Salt Lake City: University of Utah Press, 1905. Studies and Docu ments , No. 27. Lagrange, M.-J. Critique textuelle, II, La critique rationnelle. Paris: J. Gabalda, 1935

308
L a e s e r r e , p . "un nouveau m a n u s c r i t de l ' i v a n g i l e de J e a n , " Revue de TheoloKle e t de P h i l o s o p h i c , VI (1957) Lyon, Robert W. rtA Re-Examination of Codex Ephraeml R e s c r i p t u s , " New Testament S t u d i e s , V (1958/59) 260-272. M a r t i n i , C. M. "La q u e s t i o n e d e l c a r a t t e r e r e c e n s l o n a l e d e l t e s t o lucano d e l c o d i c e a l i a l u c e p a p i r o Bodmer XIV." Unpublished d i s s e r t a t i o n . Fontificio 1 s t B l b l i c o , 1965. Massaux, E . " t a t a c t u e l de l a c r i t i q u e t e x t u e l l e du Nouveau T e s t a m e n t , " Nouvelle Revue de T h ^ o l o g i e , LXXV ( 1 9 5 3 ) , 703-726^ McKeile, A. H. An I n t r o d u c t i o n t o t h e Study of t h e New Testament. Second e d i t i o n , r e v . C. 5 . C. W i l l i a a s . Oxford: At t h e Clarendon P r e s s , 1953. Henoud, P h i l i p p e - . "Papyrus Bodmer XTf-XV e t XVII," Revue de Theologle e t de P h i l o s o p h i e , XII ( 1 9 6 2 ) , 107-116. Metzger, Bruce H. "The Bodmer Papyrus of Luke and J o h n , " E x p o s i t o r y Times, LXXIII ( 1 9 6 2 ) , 2 0 1 - 2 0 3 . . "The Caesarean Text of t h e G o s p e l s , " J o u r n a l of B i b l i c a l L i t e r a t u r e , LXIV ( 1 9 ^ 5 ) , ^ 5 7 - ^ 9 - R e p r i n t e d i n Chapters in t h e H i s t o r y of New Testament T e x t u a l C r i t i c i s m . Grand R a p i d s : Wm. B . E e r d a a n s , 19e>3. Pp. 42-72. . " E x p l i c i t R e f e r e n c e s i n t h e Words of Origen t o V a r i a n t Readings i n New Testament fenuscripts," B i b l i c a l and P a t r i s t i c S t u d i e s i n Kiezjry of Robert P i e r c e Casey, e d . J . N. B l r d s a l l and R. W. Thomson. F r e i b u r g : Herder, 1963. Pp. 78-95. "The Lucianic Recension of t h e Greek B i b l e , " Chapters i n t h e H i s t o r y of New Testament Textual C r i t i c i s m . Grand R a p i d s : Wm. B . E e r d ^ a n s , I 9 0 3 . Fp^ 1-41. . "Seventy o r Seventy-two D i s c i p l e s ? " New Testament S T u d i e s , V (1958/59)* 2 9 9 - 3 0 6 .

309
Metzger, Bruce . The Text of the New Testa-aent, Its Transmission, Corruption and Restoration. New York: Oxford University Press, 1954. Michaels, J. Ramsey. "Some Notable Readings of Papyrus Bodmer II," The Bible Translator, VIII (1957), 150-15*. Murphy, Harold S. "Eusebius New Testament Text in the Demonstratlo Evangellca," Journal of Biblical Literature, LXXIII (193^), 162-163. Keufeld, Vernon H. The Earliest Christian Confessions. Grand Rapids: Wa. B. Eerdmane, 1963. Kevlus, Richard C. "The Use of the Definite Article with Jesus* in the Fourth Gospel," New Testament Studies,

X I I (1965/66), 81-85.
Oliver, Harold H. "Present Trends i n the Textual C r i t i c i s a of the New Testa-Dent,11 Journal of Bible ar.d Religion, xxx (1962), 308-320. Pack, Prank. "Origen's Evaluation of Textual Variants In the Greek B i b l e , " Restoration Quarterly, IV (i960), 139-146. . "A Study of Papyrus Bodmer I I , " Restoration Q u a r t e r l y , IV ( i 9 6 0 ) , 1-10. P o r t e r , Calvin L. "Papyrus Bodmer XV (P75) ar^d the Texrt of Codex Vaticanus," Journal of B i b l i c a l L i t e r a t u r e , LXXXI (1962), 363-376: . "A Textual Analysis of the E a r l i e s t ifenuscripts the Gospel of John." Unpublished doctoral d i s s e r t a t i o n . IX>ke University, 1961. P o t t e r i e , I . de l a . "Een nleuwe papyrus van het vlerde evangelie, pacyrus Bodaer I I , " BIJdraren, XVIII (1957), 117-128. Sacchi, Paolo. Alle o r i g i n ! del Nuove Testaaer.to: Saggio per l a s t o r i a d e l l a t r a d l z i o n e e la c r i ^ i c a eel zeszo. Pirenze, 19>6 Sanders, Henry A. "The Egyptian Text of the Pour Gospels and Acts," Earvard Theological Review, xxil (1933)*

77-98.

310 Sanders, Henry A. "Recent Text Studies In the New Testaaent," Anglican Theological Review, XVI (193^), 266-288. Shepherd, William P. "Recent Theories of Textual Critlcisa," Modem Philology, XXVIII (1930), 129-142. SaYignac, J. de. "Les fragments du Papyrus Bodmer II de l'ivangile Johannique," Scriptorium, XIV (i960),

333-338.
. "Le papyrus Bodmer II de l*vangile selon St. Jean," Scriptorium, XI (1957), 281-85. . "Les Papyrus Bodmer XIV et XV," Scriptorium, x v n (1963), 50-55. Scharleaann, Martin. "P66," Concordia Theological Monthly, XXVIII (1957), 573-573. Schweizer, Eduard. Ego Elmi . . . Die religlonsgeschlchtliche Kerkunft und theologiscre 5eceutung der johanneischen Blldreden, zusleich ein Seltrag zur Q\2ellenfrage aes vlerten g,yar. eel lugs. Gottingen: Vancterhoeck und Ruprecht, J.9^9
I I I I ' I I I I1III. I. . I 111 ,,, I.I. .Ill I I , . . I | * , ^ . 1 .

Smothers, Edgar R. "Papyrus Bodmer II: An Early Codex of St. John," Theological Studies, XVIII (1957), 434-441. - "Two Readings in Papyrus Bodmer II," Harvard Theological Review, LI (1958), 109-122. Souter, Alexander. "Progress In the Textual Criticism of the Gospels since Westcott and Eort," FansfieId College Essays. London: Hodder and Stoughton, 1909. r

Pp. 3^?-B64.'

Streeter, Burnett Hillman. The Four Gospels, A Study of Origins. London: KacKillan ana Co., Limited, 1936. . "Origen, K, and the Caesarean Text," Journal of Theological Studies, XXXVI (1935), 178-180. Suggs, X. J. "The use of Patristic Evidence in the Search for a Primitive New Testament Text," New Testament Studies, IV (1950), 139-147.

3
Taeker, R, V. 0. "The Chester Beatty Papyrus and the Caesarean Text of John," Harvard Theological Review, XXX (1937), 157-164. . "The Text of the Fourth Gospel Used by Orlgen in His Commentary on John," Journal of Theological Studies, XXXVII (1936), 146-135^ Taylor, Vincent. The Text of the New Testament. London: The MacMillan Company, 1961.
'* f

Teeple, Howard M. and P. Allyn walker. "Notes on the Plates in Papyrus Bodmer II," Journal of Biblical Literature, LXXVIII (1958), 148-152. Torrey, Charles C. "The Origin of the 'Western' Text." Documents of the Primitive Church. New York, 1941. Pp. 112-148. Turner, C. H. "Marcan Usage: Notes, Critical and Exegetical, on the Second Gospel," Journal of Theological Studies, XXV (1923/24), 377-386; XXVI (1924/25), 12-20, 145-156> 225-240, 337-346; XXVII (1925/26), 58-62; XXVIII (1926/27), 9-30, 349-362. . "A Textual Commentary on Mark 1." Journal of Theological Studies, XXVIII (1926/27), 145-158. . "Western Readings in the Second Half of St. Wark's Gospel." Journal of Theological Studies, xxix (1927/28), 17W. Turner, Nigel. Syntax. Vol. Ill of A Grammar of New Testament Greek, by James Hope Moulton. EdinDurgh: T. & T. Clark, 1963. Vaganay, Leo. An Introduction to the Textual Criticism of the New Testament, trans. B. V. Miller. London: Sands and Company, Limited, 1937. Voobus, Arthur. Early Versions of the New Testament. Stockholm: Estonian Theological Society in Exile, 1954. Wordsworth, W. A. "The Bodmer Papyrue and the Prologue of St. John's Gospel," Novum Testamentum, II (1957)* 1-7

3Xd

Zimmerman, Heinrich. "Papyrus Bodmer II und seine Bedeutung fur die Textgeschichte dee JohannesEvangeliuras," Biblische Zeltschrlft, II (1958), 214-243. Zuntz, Gunther. The Te^^ o f t n e Epistlesf A Disquisition Upon the Corpus Paulinunu London: The British Academy, 1953